|
7451
|
Fullstack python Calculator by @TheTank87 | Starda Fullstack python Calculator by @TheTank87 | Stardance...
|
TheTank87
|
1
|
2.0
|
0
|
[{"ship_num": 1, "ts": "2026-0 [{"ship_num": 1, "ts": "2026-06-10T19:15:22Z", "hours": 2.0, "multiplier": null, "payout": 0, "blessing": "neutral", "text": "Made simple python calculator script that can run by running the executable.", "repo": "https://github.com/TheTank87/Ai-Python-Calculator/tree/master", "demo": "https://github.com/TheTank87/Ai-Python-Calculator/releases/tag/executable"}]...
|
[{"post_id": 2814, "ts": "2026 [{"post_id": 2814, "ts": "2026-06-05T17:14:58Z", "hours": 1.7386, "text": "Created a python scientific calculator, still working to make it graph functions."}]...
|
Title: Fullstack python Calculator by @TheTank87 | Title: Fullstack python Calculator by @TheTank87 | Stardance
Devlogs:
Created a python scientific calculator, still working to make it graph functions.
Ships:
Made simple python calculator script that can run by running the executable....
|
1793249566.4610515
|
1793335966.4610515
|
|
7452
|
first ever slack bot by @kote5683 | Stardance
|
kote5683
|
0
|
2.1
|
0
|
[]
|
[{"post_id": 7085, "ts": "2026 [{"post_id": 7085, "ts": "2026-06-11T21:17:30Z", "hours": 0.5014, "text": "spent literally forever trying to get my nest container working, im starting from scratch now. if i cant figure this out im literally going to abandon the entire project. ive tried everything, i cant take it any more đ"}, {"post_id": 2831, "ts": "2026-06-05T17:43:01Z", "hours": 0.6189, "text": "added some more commands and uploaded it to GitHub (which took me far too long)"}, {"post_id": 2482, "ts": "2026-06-05T00:09:09Z", "hours": 0.9756, "text": "finished first ver.; currently awaiting a nest"}]...
|
Title: first ever slack bot by @kote5683 | Stardan Title: first ever slack bot by @kote5683 | Stardance
Devlogs:
finished first ver.; currently awaiting a nest
added some more commands and uploaded it to GitHub (which took me far too long)
spent literally forever trying to get my nest container working, im starting from scratch now. if i cant figure this out im literally going to abandon the entire project. ive tried everything, i cant take it any more đ...
|
1792727376.4262583
|
1793332176.4262583
|
|
7453
|
DeckController by @aloyak | Stardance
|
aloyak
|
1
|
8.0
|
102
|
[{"ship_num": 1, "ts": "2026-0 [{"ship_num": 1, "ts": "2026-07-15T15:48:01Z", "hours": 8.0, "multiplier": 12.39, "payout": 102, "blessing": "neutral", "text": "This is a two-sided system that allows you to use a steamdeck as a normal usb controller. i used to play a lot of elden ring and dark souls on my steam deck, but when i was home i wanted to play with better resolution and fps, but i realized i didnt have a controller, so i made this! hope you like it check out the yt video!!! https://www.youtube.com/watch?v=EQWmOSJWbrk", "repo": "https://github.com/aloyak/deckcontroller", "demo": "https://youtu.be/EQWmOSJWbrk"}]...
|
[{"post_id": 28940, "ts": "202 [{"post_id": 28940, "ts": "2026-07-15T15:41:55Z", "hours": 0.6603, "text": "Been working on fixing a couple things before shipping, the problem was all with windows and its super wierd api. most of these changes were done long ago but i forgot to devlog them lol anyways, ive also published a demo video to see how it works if for some reason you dont own a steamdeck. i know this is not the most quality post but im a bit tired from all the work ive been doing on other projects. so, in a way, this project is completely finished!"}, {"post_id": 4806, "ts": "2026-06-08T09:38:19Z", "hours": 3.3789, "text": "Iâve created a really nice UI with a simple controller layout that allows you to monitor the input from the steamdeck with a fancy imgui interface and also tested the performance a bit and its honestly pretty good. I managed to get around 8ms latency with 20ms spikes which is pretty good, also battery life is around 7 hours!! It currently only works on linux as windows is trying so hard to make my life suck, so ill be adding windows support later on. It currently uses a custom driver with uinput that creates a generic virtual Xbox 360 controller that is automatically recognized by my pc and also by steam so thats great! The process to make this work is pretty simple: Add the steam deck side to steam as a non-steam game and make it use proton compatibility layer Run the PC side Run the steamdeck side Thatâs it, the PC side recognizes the steamdeck instantly and thereâs no setup required at all from that point on. I would also like to add some sort of usb support but i think that will be a lot harder that im expecting :))"}, {"post_id": 4137, "ts": "2026-06-07T10:34:16Z", "hours": 2.0344, "text": "Huge progress!! Wifi connection is working nicely. Right now the best way to do it is by connecting the steamdeck to your pc with a hotspot and the latency is actually pretty nice! Theres still some small bugs but the setup is stupid simple so thats great. You just open the app on the steamdeck, and then the pc finds it right away I now should start developing the drivers to make the pc recognize this program as a generic controller but FIRST, Im going to make the UI of the desktopâs app fancy because its cool and fun to do. The idea is to fully recreate the steam controller based on an actual steamdeck, so you have all the features of the new steam controller, although its obvious that some stuff will be missing (like haptics, but Iâll see what I can do about that). And well, I also should research a bit how to make this through USB as well, but that will be done in a long time YAY"}, {"post_id": 3438, "ts": "2026-06-06T11:40:47Z", "hours": 2.1561, "text": "First ever post on stardance! Iâm making a program that allows you to use your steamdeck as a generic pc controller. Iâm sort of learning how to do this in the process so I canât really showcase any future features but so far its going great! I created a simple program using SDL3 in C++, being my first time updating from SDL2 to SDL3, but not much changed honestly (at least that I noticed) that reads the steamdecks controller input and saves it to a light struct that will later be sent using either wifi or usb to a target pc. What Iâm most proud of so far is how I managed to get a simple working environment to develop and test code on the deck quickly. I code on my computer and with the press of a button my mouse and keyboard now control the steamdeck, also being able to quickly send and receive files to test new builds etc⌠Thanks to KDE Connect and a few other tricks I will know start researching how to make any sort of wifi connection reliable enough, wish me luck! I also created a simple logo using gimp (probably not the last version, hopefully)"}]...
|
Title: DeckController by @aloyak | Stardance
Devl Title: DeckController by @aloyak | Stardance
Devlogs:
First ever post on stardance! Iâm making a program that allows you to use your steamdeck as a generic pc controller. Iâm sort of learning how to do this in the process so I canât really showcase any future features but so far its going great! I created a simple program using SDL3 in C++, being my first time updating from SDL2 to SDL3, but not much changed honestly (at least that I noticed) that reads the steamdecks controller input and saves it to a light struct that will later be sent using either wifi or usb to a target pc. What Iâm most proud of so far is how I managed to get a simple working environment to develop and test code on the deck quickly. I code on my computer and with the press of a button my mouse and keyboard now control the steamdeck, also being able to quickly send and receive files to test new builds etc⌠Thanks to KDE Connect and a few other tricks I will know start researching how to make any sort of wifi connection reliable enough, wish me luck! I also created a simple logo using gimp (probably not the last version, hopefully)
Huge progress!! Wifi connection is working nicely. Right now the best way to do it is by connecting the steamdeck to your pc with a hotspot and the latency is actually pretty nice! Theres still some small bugs but the setup is stupid simple so thats great. You just open the app on the steamdeck, and then the pc finds it right away I now should start developing the drivers to make the pc recognize this program as a generic controller but FIRST, Im going to make the UI of the desktopâs app fancy because its cool and fun to do. The idea is to fully recreate the steam controller based on an actual steamdeck, so you have all the features of the new steam controller, although its obvious that some stuff will be missing (like haptics, but Iâll see what I can do about that). And well, I also should research a bit how to make this through USB as well, but that will be done in a long time YAY
Iâve created a really nice UI with a simple controller layout that allows you to monitor the input from the steamdeck with a fancy imgui interface and also tested the performance a bit and its honestly pretty good. I managed to get around 8ms latency with 20ms spikes which is pretty good, also battery life is around 7 hours!! It currently only works on linux as windows is trying so hard to make my life suck, so ill be adding windows support later on. It currently uses a custom driver with uinput that creates a generic virtual Xbox 360 controller that is automatically recognized by my pc and also by steam so thats great! The process to make this work is pretty simple: Add the steam deck side to steam as a non-steam game and make it use proton compatibility layer Run the PC side Run the steamdeck side Thatâs it, the PC side recognizes the steamdeck instantly and thereâs no setup required at all from that point on. I would also like to add some sort of usb support but i think that will be a lot harder that im expecting :))
Been working on fixing a couple things before shipping, the problem was all with windows and its super wierd api. most of these changes were done long ago but i forgot to devlog them lol anyways, ive also published a demo video to see how it works if for some reason you dont own a steamdeck. i know this is not the most quality post but im a bit tired from all the work ive been doing on other projects. so, in a way, this project is completely finished!
Ships:
This is a two-sided system that allows you to use a steamdeck as a normal usb controller. i used to play a lot of elden ring and dark souls on my steam deck, but when i was home i wanted to play with better resolution and fps, but i realized i didnt have a controller, so i made this! hope you like it check out the yt video!!! https://www.youtube.com/watch?v=EQWmOSJWbrk...
|
1793289210.2094429
|
1793375610.2094429
|
|
7454
|
Dashboard by @TX9394 | Stardance
|
TX9394
|
1
|
71.0
|
0
|
[{"ship_num": 1, "ts": "2026-1 [{"ship_num": 1, "ts": "2026-10-30T13:11:42Z", "hours": 71.0, "multiplier": 8.35, "payout": 0, "blessing": "neutral", "text": "This is my first time shipping a project at stardance. I just made an easy dashboard, which shows you your email and appointments and stuff all in your browser. You can also add your google account, so youâre able to see your gmails on the dashboard, but you will see a warning, that the website is not approved. Just proceed, I just didnât verify.", "repo": "https://github.com/sebi-ai/dashboard", "demo": "https://dashboard.tx9394.cc"}]...
|
[{"post_id": 84843, "ts": "202 [{"post_id": 84843, "ts": "2026-10-30T13:08:25Z", "hours": 5.5572, "text": "Just fixed some bugs, everything works now. The dashboard is now ready for shiping. When you want to connect your google account, a warning will appear, which tells you the website is not safe, because I didnât approve it with google safety standards. Just press the more options button and then continue to the dashboard."}, {"post_id": 77365, "ts": "2026-10-06T17:52:02Z", "hours": 10.7172, "text": "Devlog 38 I used the whole time to get the python dashboard on the website. Streamlit helped a lot and now I finally have a version which works (sometimes). There are still some errors with API keys and the settings being transferred correctly on the website."}, {"post_id": 39355, "ts": "2026-08-02T08:47:49Z", "hours": 0.6644, "text": "Devlog 37 I didnât do much, I just added some functions which got lost by my project reset. Somehow, the buttons on the website didnât work also, so I fixed that."}, {"post_id": 38727, "ts": "2026-08-01T08:48:33Z", "hours": 10.3597, "text": "Devlog 36 Thatâs a problem right now, because I did exactly nothing. I added the function, that your google account that you added, is saved in the json file. But because of some stupid reason, the website didnât find the json file anymore as soon as the google account was connected. I tried every humanly possible ways to fix that, I didnât have any success. Then I asked Claude Code, who also didnât know what to do, it didnât work. I now deleted all of the changes and brought it back to the last commit on github were everything still works. Next step is to find a solution, so the google account gets saved in the json file, so the dashboard can show your calender entries and stuff. I know itâs not a good idea to log this time right now, but I didnât find a way to âdeleteâ time, and I also just donât want to start this project over again."}, {"post_id": 29668, "ts": "2026-07-16T18:51:32Z", "hours": 1.6069, "text": "Devlog 35 1. Dashboard design The curved edges didnât want to work propably, so I made them not curved again Fixed some color issues, because at some specific color like dark red the contrast was trash 2. Widgets The weather widget works again now, that was just a small fx I am currently having a lot of trouble with the calender widget, because I always wondered why the appointments arenât shown probably on the dashboard. Now I found out why: Because I am saving the google account exactly nowhere, wo the python dashboard doesnât have access to the account information. I donât know yet, how I want to save an entire google account in a json file, but I am sure Iâll find I way to do it (hopefully)"}, {"post_id": 29077, "ts": "2026-07-15T18:48:19Z", "hours": 0.8064, "text": "Devlog 34 1. Google Google still wasnât able to verify my app, so itâs still just avaible to specific test users The calender/notification date is also not shown properly yet, there is some problem with the json file, I still wasnât able to fix this 2. Python dashboard The corners of the dashboard widgets are somehow not as curved as I want them to be, and the resolution is also trash some how"}, {"post_id": 28471, "ts": "2026-07-14T18:37:42Z", "hours": 3.8711, "text": "Devlog 33 1. Google On the Python dashboard, neither your google calender nor your notifications are shown proberly. I still have to fix this I tried verifying my app with google, so itâs public and anyone can use it and I donât have to declare you as test user for you to be able to use the website and show your notifications and stuff 2. Dashboard Design The edges of the python dashboard are curved now The dashboard also works now with really dark of really light color and the contrast isnât shit anymore"}, {"post_id": 25460, "ts": "2026-07-09T18:58:48Z", "hours": 1.3867, "text": "Devlog 32 Dashboard Design If you choose a custom color now, not just the text and the lines have that color, but the whole website. It was pretty complicated to be sure, that the contrast is neither at really light nor at really dark color ass."}, {"post_id": 24893, "ts": "2026-07-08T18:57:39Z", "hours": 2.0447, "text": "#Devlog 31 1. Getting the website ready There was again a problem with the google services, I submitted my website now for verification. Because of some reason you werenât able to scroll in the settings section, I fixed that 2. Python dashboard I started with the dashboard, I pretty much redesigned it, it looks different to the old version. The google services like calender or gmail donât work yet and the color isnât as I want it yet. But itâs a good start. I again had lots of help by AI here, because as I already wrote, my python knowledge is pretty ass."}, {"post_id": 23650, "ts": "2026-07-06T18:55:45Z", "hours": 1.1494, "text": "#Devlog 30 1. Google Now you finally can add your google account to the website That just works if you are on localhost, on server, it doesnât work yet, donât really now why The problem was, that I didnât give google any authorized javascript origins, which leads to the problem, that the origin is authorized 2. API Safety Because of the new json file with my google api key and everything, the json files were accesible for anyone in internet, I did also fix that"}, {"post_id": 22978, "ts": "2026-07-05T17:43:21Z", "hours": 1.3053, "text": "Devlog 29 1. Google I now created a completly new project at google claude, because somehow the website always said, the client was deleted. It still says that I donât now what to do. Iâll just wait now, maybe it doesnât load properly With the new google project, I added the new secret and the new google id to the .env I donât have an .env file for the google secret and id anymore, I now use the json file by google 2. Fixing design issues Somehow, one checkbox was always the wrong size. Thatâs because the text was too big. I now made the text smaller, so the checkbox has itâs proper size now Now, if you select the checkbox for preset theme, you canât choose a custom color any more. You also canât select a preset theme, if the checkbox for custom color is checked"}, {"post_id": 20991, "ts": "2026-07-02T18:49:36Z", "hours": 0.2781, "text": "I think the website is actually ready for the python script now."}, {"post_id": 20024, "ts": "2026-07-01T14:59:46Z", "hours": 0.4256, "text": "I tried to fix the issue with the about div, so it doesnât move. Wasnât successful again."}, {"post_id": 19270, "ts": "2026-06-30T15:12:26Z", "hours": 1.1172, "text": "I spent the whole fucking time trying to tell this stupid piece of shit to not fucking move. It did ignore me of course all the time. Whatever, I donât want this anymore."}, {"post_id": 19108, "ts": "2026-06-30T10:44:59Z", "hours": 0.7, "text": "Just fixed some pretty stupid design bugs"}, {"post_id": 18389, "ts": "2026-06-29T15:02:42Z", "hours": 1.505, "text": "I added a new menu design, so the menu hovers open now."}, {"post_id": 17767, "ts": "2026-06-28T18:15:35Z", "hours": 0.3, "text": "I now added the functinality for the start dashboard button"}, {"post_id": 17032, "ts": "2026-06-27T17:19:42Z", "hours": 2.6183, "text": "I now started with the actual python dashboard, i am not happy with the design yet and half of the functions donât work right now, but i think itâs a pretty nice start"}, {"post_id": 16430, "ts": "2026-06-26T18:48:31Z", "hours": 1.7358, "text": "Now, you canât just get the .env file anymore. I also added the function, that if one person connects her google account, not everyone can see his appointments and everything."}, {"post_id": 15219, "ts": "2026-06-24T18:43:27Z", "hours": 1.1431, "text": "I fixed some bugs in the design If you open the windows for the notifications or something like that, the background, so the website gets blurry I added a favicon/logo"}, {"post_id": 14998, "ts": "2026-06-24T11:35:24Z", "hours": 0.34, "text": "I started with the notification widget, but you only have your emails yet."}, {"post_id": 14944, "ts": "2026-06-24T08:44:40Z", "hours": 0.9147, "text": "now you can also change your settings by pressing the change buttons, not only when ticking the checkboxes."}, {"post_id": 14558, "ts": "2026-06-23T18:40:11Z", "hours": 0.5456, "text": "I started adding the possibility to add stocks or crypto currencies"}, {"post_id": 14350, "ts": "2026-06-23T13:32:05Z", "hours": 0.4947, "text": "Now it finally shows everything from your google calender also in the dashboard. I had a lots of help by ai here."}, {"post_id": 13705, "ts": "2026-06-22T14:02:11Z", "hours": 0.8136, "text": "I now started adding the possibilty to add your google calender to the dashboard. thats a lot more difficult than I thought at first. I deleted half of my code now, thatâs why I donâthave any screenshots right now."}, {"post_id": 13160, "ts": "2026-06-21T15:11:21Z", "hours": 1.0911, "text": "I now styled the hover and click animation for the new button styles. I also added the about me content and styled it."}, {"post_id": 11917, "ts": "2026-06-19T13:04:59Z", "hours": 0.2939, "text": "I redesigned the website, it has a different font and a different color now."}, {"post_id": 10808, "ts": "2026-06-17T18:45:09Z", "hours": 1.9222, "text": "Added a new font, but I have to look for another one, donât really like that one yet Now if you select the countdown to date widget, there opens a window, in which you can select the date. There were some problems with the IP location, so I tried to fix them, didnât have that much success."}, {"post_id": 10128, "ts": "2026-06-16T18:47:46Z", "hours": 1.8225, "text": "I completly redesigned the notifications, because the they are declarated as âsuccessâ so the browser styles them as âsuccessâ but i donât want this, because this doesnât match the design of the website. Every section like settings or home has its own html file now for more structure and less chaos Fixed some JS and CSS bugs"}, {"post_id": 9252, "ts": "2026-06-15T15:13:19Z", "hours": 1.4561, "text": "Now you canât save your settings without filling out everything"}, {"post_id": 7613, "ts": "2026-06-12T18:42:55Z", "hours": 1.7892, "text": "Finally the settings saving works properly now. I also started writing my readme"}, {"post_id": 6986, "ts": "2026-06-11T18:43:54Z", "hours": 2.2928, "text": "I started doing some backend, so if you change the settings, they also get saved in a json file. I also added one new tab to the settings, were you can choose if you want your dashboard to have an preset theme or you can even choose your own color."}, {"post_id": 6366, "ts": "2026-06-10T18:38:50Z", "hours": 3.6539, "text": "I started adding the first settings and styling them."}, {"post_id": 5806, "ts": "2026-06-09T18:51:07Z", "hours": 0.7542, "text": "I now started with CSS, the buttons were a good start."}, {"post_id": 5607, "ts": "2026-06-09T13:56:14Z", "hours": 1.3825, "text": "I had massive problems with Git, because of the venv. Now I deleted it, everything works perfectly now. Also added css."}, {"post_id": 4901, "ts": "2026-06-08T14:02:58Z", "hours": 0.7464, "text": "Mainly fixed bugs, also added some css"}, {"post_id": 4804, "ts": "2026-06-08T09:33:43Z", "hours": 0.3194, "text": "Had some time in school, so I started adding css and javascript. Struggeled a lot with JavaScript, donât know much about it."}, {"post_id": 4412, "ts": "2026-06-07T18:48:49Z", "hours": 0.6278, "text": "As planned, I am coding the website by myself, so itâs a lot slow, but the structure is ready, just need some css."}, {"post_id": 2662, "ts": "2026-06-05T10:26:19Z", "hours": 0.9458, "text": "Started with the python dashboard, there will get some more information here, but its a nice start. I did that with a lot of help of AI, because I am not good yet at python, but I was able to do the debugging. I am trying do to the website by myself."}]...
|
Title: Dashboard by @TX9394 | Stardance
Devlogs:
Title: Dashboard by @TX9394 | Stardance
Devlogs:
Started with the python dashboard, there will get some more information here, but its a nice start. I did that with a lot of help of AI, because I am not good yet at python, but I was able to do the debugging. I am trying do to the website by myself.
As planned, I am coding the website by myself, so itâs a lot slow, but the structure is ready, just need some css.
Had some time in school, so I started adding css and javascript. Struggeled a lot with JavaScript, donât know much about it.
Mainly fixed bugs, also added some css
I had massive problems with Git, because of the venv. Now I deleted it, everything works perfectly now. Also added css.
I now started with CSS, the buttons were a good start.
I started adding the first settings and styling them.
I started doing some backend, so if you change the settings, they also get saved in a json file. I also added one new tab to the settings, were you can choose if you want your dashboard to have an preset theme or you can even choose your own color.
Finally the settings saving works properly now. I also started writing my readme
Now you canât save your settings without filling out everything
I completly redesigned the notifications, because the they are declarated as âsuccessâ so the browser styles them as âsuccessâ but i donât want this, because this doesnât match the design of the website. Every section like settings or home has its own html file now for more structure and less chaos Fixed some JS and CSS bugs
Added a new font, but I have to look for another one, donât really like that one yet Now if you select the countdown to date widget, there opens a window, in which you can select the date. There were some problems with the IP location, so I tried to fix them, didnât have that much success.
I redesigned the website, it has a different font and a different color now.
I now styled the hover and click animation for the new button styles. I also added the about me content and styled it.
I now started adding the possibilty to add your google calender to the dashboard. thats a lot more difficult than I thought at first. I deleted half of my code now, thatâs why I donâthave any screenshots right now.
Now it finally shows everything from your google calender also in the dashboard. I had a lots of help by ai here.
I started adding the possibility to add stocks or crypto currencies
now you can also change your settings by pressing the change buttons, not only when ticking the checkboxes.
I started with the notification widget, but you only have your emails yet.
I fixed some bugs in the design If you open the windows for the notifications or something like that, the background, so the website gets blurry I added a favicon/logo
Now, you canât just get the .env file anymore. I also added the function, that if one person connects her google account, not everyone can see his appointments and everything.
I now started with the actual python dashboard, i am not happy with the design yet and half of the functions donât work right now, but i think itâs a pretty nice start
I now added the functinality for the start dashboard button
I added a new menu design, so the menu hovers open now.
Just fixed some pretty stupid design bugs
I spent the whole fucking time trying to tell this stupid piece of shit to not fucking move. It did ignore me of course all the time. Whatever, I donât want this anymore.
I tried to fix the issue with the about div, so it doesnât move. Wasnât successful again.
I think the website is actually ready for the python script now.
Devlog 29 1. Google I now created a completly new project at google claude, because somehow the website always said, the client was deleted. It still says that I donât now what to do. Iâll just wait now, maybe it doesnât load properly With the new google project, I added the new secret and the new google id to the .env I donât have an .env file for the google secret and id anymore, I now use the json file by google 2. Fixing design issues Somehow, one checkbox was always the wrong size. Thatâs because the text was too big. I now made the text smaller, so the checkbox has itâs proper size now Now, if you select the checkbox for preset theme, you canât choose a custom color any more. You also canât select a preset theme, if the checkbox for custom color is checked
#Devlog 30 1. Google Now you finally can add your google account to the website That just works if you are on localhost, on server, it doesnât work yet, donât really now why The problem was, that I didnât give google any authorized javascript origins, which leads to the problem, that the origin is authorized 2. API Safety Because of the new json file with my google api key and everything, the json files were accesible for anyone in internet, I did also fix that
#Devlog 31 1. Getting the website ready There was again a problem with the google services, I submitted my website now for verification. Because of some reason you werenât able to scroll in the settings section, I fixed that 2. Python dashboard I started with the dashboard, I pretty much redesigned it, it looks different to the old version. The google services like calender or gmail donât work yet and the color isnât as I want it yet. But itâs a good start. I again had lots of help by AI here, because as I already wrote, my python knowledge is pretty ass.
Devlog 32 Dashboard Design If you choose a custom color now, not just the text and the lines have that color, but the whole website. It was pretty complicated to be sure, that the contrast is neither at really light nor at really dark color ass.
Devlog 33 1. Google On the Python dashboard, neither your google calender nor your notifications are shown proberly. I still have to fix this I tried verifying my app with google, so itâs public and anyone can use it and I donât have to declare you as test user for you to be able to use the website and show your notifications and stuff 2. Dashboard Design The edges of the python dashboard are curved now The dashboard also works now with really dark of really light color and the contrast isnât shit anymore
Devlog 34 1. Google Google still wasnât able to verify my app, so itâs still just avaible to specific test users The calender/notification date is also not shown properly yet, there is some problem with the json file, I still wasnât able to fix this 2. Python dashboard The corners of the dashboard widgets are somehow not as curved as I want them to be, and the resolution is also trash some how
Devlog 35 1. Dashboard design The curved edges didnât want to work propably, so I made them not curved again Fixed some color issues, because at some specific color like dark red the contrast was trash 2. Widgets The weather widget works again now, that was just a small fx I am currently having a lot of trouble with the calender widget, because I always wondered why the appointments arenât shown probably on the dashboard. Now I found out why: Because I am saving the google account exactly nowhere, wo the python dashboard doesnât have access to the account information. I donât know yet, how I want to save an entire google account in a json file, but I am sure Iâll find I way to do it (hopefully)
Devlog 36 Thatâs a problem right now, because I did exactly nothing. I added the function, that your google account that you added, is saved in the json file. But because of some stupid reason, the website didnât find the json file anymore as soon as the google account was connected. I tried every humanly possible ways to fix that, I didnât have any success. Then I asked Claude Code, who also didnât know what to do, it didnât work. I now deleted all of the changes and brought it back to the last commit on github were everything still works. Next step is to find a solution, so the google account gets saved in the json file, so the dashboard can show your calender entries and stuff. I know itâs not a good idea to log this time right now, but I didnât find a way to âdeleteâ time, and I also just donât want to start this project over again.
Devlog 37 I didnât do much, I just added some functions which got lost by my project reset. Somehow, the buttons on the website didnât work also, so I fixed that.
Devlog 38 I used the whole time to get the python dashboard on the website. Streamlit helped a lot and now I finally have a version which works (sometimes). There are still some errors with API keys and the settings being transferred correctly on the website.
Just fixed some bugs, everything works now. The dashboard is now ready for shiping. When you want to connect your google account, a warning will appear, which tells you the website is not safe, because I didnât approve it with google safety standards. Just press the more options button and then continue to the dashboard.
Ships:
This is my first time shipping a project at stardance. I just made an easy dashboard, which shows you your email and appointments and stuff all in your browser. You can also add your google account, so youâre able to see your gmails on the dashboard, but you will see a warning, that the website is not approved. Just proceed, I just didnât verify....
|
1793407839.3988752
|
1793494239.3988752
|
|
7455
|
Hyprland + Quickshell Dots by @lukas | Stardance
|
lukas
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Hyprland + Quickshell Dots by @lukas | Star Title: Hyprland + Quickshell Dots by @lukas | Stardance...
|
1792756217.8359327
|
1793361017.8359327
|
|
7456
|
WMCF (What's My Carbon Footprint?) by @maka | WMCF (What's My Carbon Footprint?) by @maka | Stardance...
|
maka
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: WMCF (What's My Carbon Footprint?) by @ Title: WMCF (What's My Carbon Footprint?) by @maka | Stardance...
|
1792788897.210238
|
1793393697.210238
|
|
7457
|
x by @Nancy | Stardance
|
Nancy
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: x by @Nancy | Stardance
|
1792734522.8072312
|
1793339322.8072312
|
|
7458
|
My first slack bot! by @solarlunarr | Stardance
|
solarlunarr
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My first slack bot! by @solarlunarr | Stard Title: My first slack bot! by @solarlunarr | Stardance...
|
1792731032.1155777
|
1793335832.1155777
|
|
7459
|
Stardew Valley Pebble Watch Face by @Loekman | Sta Stardew Valley Pebble Watch Face by @Loekman | Stardance...
|
Loekman
|
1
|
9.0
|
0
|
[{"ship_num": 1, "ts": "2026-0 [{"ship_num": 1, "ts": "2026-06-11T14:26:13Z", "hours": 9.0, "multiplier": null, "payout": 0, "blessing": "neutral", "text": "I created a watch face for the Pebble based on the Stardew Valley clock. Since this was my first time developing something for a Pebble, it was quite challenging. However, it is very rewarding to know that I have a custom watch face ready to test as soon as I receive my watch.", "repo": "https://github.com/Loekiboy/stardew_valley_pebble", "demo": "https://apps.repebble.com/stardew-valley-watchface_954dddca9a9c44f0ab1c3ff8"}]...
|
[{"post_id": 6276, "ts": "2026 [{"post_id": 6276, "ts": "2026-06-10T15:35:54Z", "hours": 0.3411, "text": "I Upoaded the watchface to the Pebble appstore you can now view and download it here: https://apps.repebble.com/954dddca9a9c44f0ab1c3ff8"}, {"post_id": 5705, "ts": "2026-06-09T16:31:23Z", "hours": 1.6336, "text": "I added a step counter. I edited the image so there stands an âSâ instead of a âGâ. For the step counter, I used a new font. It was really hard to get the numbers perfectly into the slots of the counter."}, {"post_id": 5012, "ts": "2026-06-08T17:37:07Z", "hours": 1.1819, "text": "I added the weather badge, which changes depending on the current weather. I also built a feature that automatically pops up a festival badge on special days like Christmas, New Year, or Easter! And I added a custom logo to the watchface."}, {"post_id": 4929, "ts": "2026-06-08T15:10:56Z", "hours": 0.3181, "text": "I added the season badge, the image changes depending on the season."}, {"post_id": 4324, "ts": "2026-06-07T16:16:08Z", "hours": 0.7092, "text": "I have added the dial to the watch face. The dial points downwards at 06:00 and slowly moves upwards until it reaches 00:00. It then remains upright until 06:00, at which point the dial jumps back down."}, {"post_id": 4261, "ts": "2026-06-07T14:51:44Z", "hours": 1.0278, "text": "I added the clock part to the watch face and adjusted the positioning of the time and date accordingly. And I updated the background image, the original gradient did not render well on the pebble, screen so I used this site https://loek.oerlemans.tv/pebble_images/ to generate a gradient that is optimized for the Pebble display."}, {"post_id": 4167, "ts": "2026-06-07T11:51:22Z", "hours": 1.9139, "text": "I added a background to the watch face. I tried to add the background for a very long time but I couldnât get it to work. Ultimately, I remade the project in C instead of JS, and then it worked."}, {"post_id": 3402, "ts": "2026-06-06T09:53:57Z", "hours": 1.6928, "text": "I changed the standard font to the Stardew Valley pixel art font. I also tried to change the background, but everything I tried didnât work, so I will continue with that later."}]...
|
Title: Stardew Valley Pebble Watch Face by @Loekma Title: Stardew Valley Pebble Watch Face by @Loekman | Stardance
Devlogs:
I changed the standard font to the Stardew Valley pixel art font. I also tried to change the background, but everything I tried didnât work, so I will continue with that later.
I added a background to the watch face. I tried to add the background for a very long time but I couldnât get it to work. Ultimately, I remade the project in C instead of JS, and then it worked.
I added the clock part to the watch face and adjusted the positioning of the time and date accordingly. And I updated the background image, the original gradient did not render well on the pebble, screen so I used this site https://loek.oerlemans.tv/pebble_images/ to generate a gradient that is optimized for the Pebble display.
I have added the dial to the watch face. The dial points downwards at 06:00 and slowly moves upwards until it reaches 00:00. It then remains upright until 06:00, at which point the dial jumps back down.
I added the season badge, the image changes depending on the season.
I added the weather badge, which changes depending on the current weather. I also built a feature that automatically pops up a festival badge on special days like Christmas, New Year, or Easter! And I added a custom logo to the watchface.
I added a step counter. I edited the image so there stands an âSâ instead of a âGâ. For the step counter, I used a new font. It was really hard to get the numbers perfectly into the slots of the counter.
I Upoaded the watchface to the Pebble appstore you can now view and download it here: https://apps.repebble.com/954dddca9a9c44f0ab1c3ff8
Ships:
I created a watch face for the Pebble based on the Stardew Valley clock. Since this was my first time developing something for a Pebble, it was quite challenging. However, it is very rewarding to know that I have a custom watch face ready to test as soon as I receive my watch....
|
1793267416.5703776
|
1793353816.5703776
|
|
7460
|
Halo themed HackPad by @Hridayesh | Stardance
|
Hridayesh
|
0
|
2.94
|
0
|
[]
|
[{"post_id": 43400, "ts": "202 [{"post_id": 43400, "ts": "2026-08-08T16:55:47Z", "hours": 1.35, "text": "Dev Log 2 : Today I decided to finally start making the CAD model on Onshape and it has taken a lot of time since I have not used Onshape before this project. It doesnât really look that great yet but the biggest problem I think will be making the Plate fit my PCB design as I currently donât know how to solve that issue"}, {"post_id": 31399, "ts": "2026-07-19T18:48:50Z", "hours": 1.5922, "text": "Dev log 1: This is technically my second day working on this project but due to some stupid mistake I made with Lookout, I cant use the recording I made and log the 1+ Hours. So in these 2 days I finished the entire PCB design and now I will work on the CAD model ."}]...
|
Title: Halo themed HackPad by @Hridayesh | Stardan Title: Halo themed HackPad by @Hridayesh | Stardance
Devlogs:
Dev log 1: This is technically my second day working on this project but due to some stupid mistake I made with Lookout, I cant use the recording I made and log the 1+ Hours. So in these 2 days I finished the entire PCB design and now I will work on the CAD model .
Dev Log 2 : Today I decided to finally start making the CAD model on Onshape and it has taken a lot of time since I have not used Onshape before this project. It doesnât really look that great yet but the biggest problem I think will be making the Plate fit my PCB design as I currently donât know how to solve that issue...
|
1792723829.0246975
|
1793328629.0246975
|
|
7461
|
My awesome macropad! by @colin_brahan | Stardance
|
colin_brahan
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @colin_brahan | Sta Title: My awesome macropad! by @colin_brahan | Stardance...
|
1793372071.291675
|
1793976871.291675
|
|
7462
|
Photography Portfolio by @Elied | Stardance
|
Elied
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Photography Portfolio by @Elied | Stardance
|
1793267473.0699756
|
1793872273.0699756
|
|
7463
|
PixelOrbit by @leotsio | Stardance
|
leotsio
|
0
|
22.49
|
0
|
[]
|
[{"post_id": 14941, "ts": "202 [{"post_id": 14941, "ts": "2026-06-24T08:41:13Z", "hours": 3.5547, "text": "đž PixelOrbit - Devlog #4 (Perseverance sprite & UI improvements) đ New spacecraft (kind ofâŚ) I added perseverance in the spacecraftData.json and made a sprite for it with my pixelator.py helper program (check it out in my first devlog), however i havenât made a provider for it yet. I will also add JWST and Voyager 1 , so help me decide what data should i add for these along with perseverance. UI polish (the good stuff) I didnât actually work hard on this update, but i made the frontend a bit more engaging. Ngl i have not very high css skills and i left most of the weird styling in chatgpt but the whole design is mine. First of all, i added a slight parallax effect that helped me move on from this static design. gave it some colour , since before it was like a black and white 90s photo. Another huge UI redesign was the navigator which now has a completely new look and now works like a real carousel . â ď¸ Debugging I also noticed a minor bug and had the chance to fix it. well, both side arrows triggered the same function which made the UI feel broken because no matter to which side you wanted to change spacecraft, it always slid to the right. Resources check it out at: PixelOrbit The repo: LeoTsio/PixelOrbit"}, {"post_id": 11862, "ts": "2026-06-19T09:44:25Z", "hours": 8.6467, "text": "đž PixelOrbit â Devlog #3 (UI) The Goal PixelOrbit is a pixel-art style website that has one goal in mind: Make NASA mission data & telemetry understandable by anyone. After hours of a C++ guy trying to write javascript i finally finished the first version of the UI . have a look: PixelOrbit Making It Real This was the moment where the project actually started to feel real to me. Instead of random bits of code and half-working ideas, PixelOrbit now has an actual face: a proper landing view, the spacecraft front and center, a selector to switch between them, and a data panel that actually belongs there. What it has now a proper landing view the spacecraft front and center a selector to switch between them a data panel that feels like it actually belongs there âQUESTION I would love any feedback from all of you on any improvements, but most importantly what data would you like to see for JWST, voyager 1 and perseverance. Good options are the most unexpected ones who anyone can understand."}, {"post_id": 6390, "ts": "2026-06-10T19:03:29Z", "hours": 3.2056, "text": "đž PixelOrbit â Devlog #2 (Providers) Iâm building PixelOrbit , a retro pixel-art website that turns real NASA spacecraft telemetry into something anyone can understand. This week I finished the first working version of the project. Provider System I built a provider system where every spacecraft has its own JavaScript file ( iss.js , hubble.js , etc.) responsible for fetching and processing data. The UI doesnât need to know anything about the spacecraft itselfâit simply asks the current provider for data and displays the result. This should make adding future spacecraft incredibly easy. Progress I also reorganized the entire project structure, added a testing UI, and got live telemetry working for both the ISS and Hubble. After spending far more time fighting GitHub Pages, CORS restrictions, HTTPS issues, and random API limitations than actually coding features, I finally managed to get everything running online. đ Live website: https://leotsio.github.io/PixelOrbit/ Current Spacecraft đ°ď¸ ISS â Altitude, velocity, current location đ Hubble â Altitude, velocity, current location â Question for Everyone What data would you actually want to see for spacecraft like the ISS, Hubble, JWST, Voyager 1, or Perseverance? Iâm trying to focus on the stats that are genuinely fascinating rather than just technically available."}, {"post_id": 4900, "ts": "2026-06-08T14:02:53Z", "hours": 7.0839, "text": "đž PixelOrbit - Devlog #1 (Pixelator) Iâm building PixelOrbit , a pixel-art website that transforms real NASA spacecraft telemetry into interactive visualizations that are easier to understand. Today I completed the first milestone of the project: the Pixelator . The Pixelator takes an image from either a local file or an internet URL and automatically: Removes the background Crops the main object Converts it into pixel art Exports a transparent PNG The image below shows the first successful conversion of the Hubble Space Telescope . Most of todayâs work was spent debugging dependencies and getting the background-removal pipeline running on my 11-year-old MacBook Air, but the first version is now working and ready for the next stage of the project đ"}]...
|
Title: PixelOrbit by @leotsio | Stardance
Devlogs Title: PixelOrbit by @leotsio | Stardance
Devlogs:
đž PixelOrbit - Devlog #1 (Pixelator) Iâm building PixelOrbit , a pixel-art website that transforms real NASA spacecraft telemetry into interactive visualizations that are easier to understand. Today I completed the first milestone of the project: the Pixelator . The Pixelator takes an image from either a local file or an internet URL and automatically: Removes the background Crops the main object Converts it into pixel art Exports a transparent PNG The image below shows the first successful conversion of the Hubble Space Telescope . Most of todayâs work was spent debugging dependencies and getting the background-removal pipeline running on my 11-year-old MacBook Air, but the first version is now working and ready for the next stage of the project đ
đž PixelOrbit â Devlog #2 (Providers) Iâm building PixelOrbit , a retro pixel-art website that turns real NASA spacecraft telemetry into something anyone can understand. This week I finished the first working version of the project. Provider System I built a provider system where every spacecraft has its own JavaScript file ( iss.js , hubble.js , etc.) responsible for fetching and processing data. The UI doesnât need to know anything about the spacecraft itselfâit simply asks the current provider for data and displays the result. This should make adding future spacecraft incredibly easy. Progress I also reorganized the entire project structure, added a testing UI, and got live telemetry working for both the ISS and Hubble. After spending far more time fighting GitHub Pages, CORS restrictions, HTTPS issues, and random API limitations than actually coding features, I finally managed to get everything running online. đ Live website: https://leotsio.github.io/PixelOrbit/ Current Spacecraft đ°ď¸ ISS â Altitude, velocity, current location đ Hubble â Altitude, velocity, current location â Question for Everyone What data would you actually want to see for spacecraft like the ISS, Hubble, JWST, Voyager 1, or Perseverance? Iâm trying to focus on the stats that are genuinely fascinating rather than just technically available.
đž PixelOrbit â Devlog #3 (UI) The Goal PixelOrbit is a pixel-art style website that has one goal in mind: Make NASA mission data & telemetry understandable by anyone. After hours of a C++ guy trying to write javascript i finally finished the first version of the UI . have a look: PixelOrbit Making It Real This was the moment where the project actually started to feel real to me. Instead of random bits of code and half-working ideas, PixelOrbit now has an actual face: a proper landing view, the spacecraft front and center, a selector to switch between them, and a data panel that actually belongs there. What it has now a proper landing view the spacecraft front and center a selector to switch between them a data panel that feels like it actually belongs there âQUESTION I would love any feedback from all of you on any improvements, but most importantly what data would you like to see for JWST, voyager 1 and perseverance. Good options are the most unexpected ones who anyone can understand.
đž PixelOrbit - Devlog #4 (Perseverance sprite & UI improvements) đ New spacecraft (kind ofâŚ) I added perseverance in the spacecraftData.json and made a sprite for it with my pixelator.py helper program (check it out in my first devlog), however i havenât made a provider for it yet. I will also add JWST and Voyager 1 , so help me decide what data should i add for these along with perseverance. UI polish (the good stuff) I didnât actually work hard on this update, but i made the frontend a bit more engaging. Ngl i have not very high css skills and i left most of the weird styling in chatgpt but the whole design is mine. First of all, i added a slight parallax effect that helped me move on from this static design. gave it some colour , since before it was like a black and white 90s photo. Another huge UI redesign was the navigator which now has a completely new look and now works like a real carousel . â ď¸ Debugging I also noticed a minor bug and had the chance to fix it. well, both side arrows triggered the same function which made the UI feel broken because no matter to which side you wanted to change spacecraft, it always slid to the right. Resources check it out at: PixelOrbit The repo: LeoTsio/PixelOrbit...
|
1793260424.140681
|
1793865224.140681
|
|
7464
|
My first slack bot! by @sol | Stardance
|
sol
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My first slack bot! by @sol | Stardance
|
1793332033.84699
|
1793936833.84699
|
|
7465
|
My first slack bot! by @Wizzy | Stardance
|
Wizzy
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My first slack bot! by @Wizzy | Stardance
|
1793217180.9139748
|
1793821980.9139748
|
|
7466
|
My first slack bot! by @izza | Stardance
|
izza
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My first slack bot! by @izza | Stardance
|
1792716600.4130223
|
1793321400.4130223
|
|
7467
|
Oblisk of Malice by @Red_from1 | Stardance
|
Red_from1
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Oblisk of Malice by @Red_from1 | Stardance
|
1793029703.591606
|
1793634503.591606
|
|
7468
|
My awesome macropad! by @ARY2 | Stardance
|
ARY2
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @ARY2 | Stardance
|
1792756269.6932359
|
1793361069.6932359
|
|
7469
|
Untitled by @Wizzy | Stardance
|
Wizzy
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Untitled by @Wizzy | Stardance
|
1793325112.010806
|
1793929912.010806
|
|
7470
|
Dungeon-Adventure by @Reid | Stardance
|
Reid
|
1
|
31.66
|
286
|
[{"ship_num": 1, "ts": "2026-0 [{"ship_num": 1, "ts": "2026-07-01T22:10:55Z", "hours": 26.0, "multiplier": 11.61, "payout": 286, "blessing": "neutral", "text": "This is my first ship so I hope you all enjoy it. I made a Dungeon Adventure game that you run in your terminal or shell depending on what device you use. You can go through different dungeon levels that are randomly generated each time you enter a new one. You can also buy lifes or open chests in the shop using gold or keys that you can gather from withen the dungeon, but be careful because there are enemies that can attack you as well so keep an eye on your life and where the enemies are. Some challenges I faced while making this project were figuring out how to make the enemy movement and make it work well, I would say it was probably one of my most difficult and common problems because i had to rewrite how the enemies spawned and functioned over and over until i figured it out. I also had problems with making the small room generations withen the dungeon, because they would constantly keep generating paths that went outside the map causing a crash and i couldnt figure out why for the longest time (like most times with a problem like this it was a kinda simple solution but i still took forever to figure it out). I think my favorite part was making the random map generation because I found it fun to make and I built an algorithm to do it basically from the ground up which was exciting. I did have alot of problems with this too but I figured them out as well. This game is run in a terminal and you do need external libraries such as readchar and colorama to run the program. It comes in 2 files one is the main one you run and the other has a bunch of functions that are called from the first one or even from other functions because alot of the things need to be repeated constantly. Anyways i hope you guys enjoy this project and that it works for you (because im kinda scared that I uploaded something wrong or something to where it wont work for you guys) So have fun and let me know if anything isnt working right by commenting on one of my devlogs or contacting me in Slack (Jason Challand)", "repo": "https://github.com/GorillaReid/Dungeon-Adventure", "demo": "https://github.com/GorillaReid/Dungeon-Adventure/releases/tag/v1.0.0"}]...
|
[{"post_id": 44752, "ts": "202 [{"post_id": 44752, "ts": "2026-08-10T18:44:14Z", "hours": 5.6603, "text": "So I took a break and itâs been a while since I worked on the project but I did finally start working on it again. Iâve decided that since itâs only compatible with some Linux devices I would make a html version that can run in browsers, I did some research and found out I could use pyodide to run Python in a html/javascript environment so I started using that and basically copy pasted most my code into the new file, I have had to modify some of the code like the movement and coloration to work with JavaScript instead. I also decided to make going into the different menus have buttons instead of pushing keys, but Iâm having a lot of trouble with getting the buttons to display at the right time and function correctly. Hopefully I can get them working right by the next devlog. (P.S. Iâm new to html and JavaScript so I donât fully know how it works and I kinda feel like buttons should be easier and Iâm just missing something, but maybe not and they really are just hard.)"}, {"post_id": 20316, "ts": "2026-07-01T21:40:51Z", "hours": 6.0642, "text": "This is probably my longest work period between devlogs, I did also take like a week break in between them to. Im also hoping i can ship the project after this I worked on making the enemy movement smoother and work better, i decided to not use threading to run multiple loops since i didnât like how it worked so i just did it semi turn based where the player moves and the enemies will then move, but its so quick the player can just hold down a button to move and it will move them multiple spaces seamlessly. I also worked on polishing the UIs like the Inventory and Shop so that they look better, they now have color and look nice. Something else i worked on was making a Info option too that tells you a little about the game I worked on making the code look cleaner and I added a bunch of comments so that anyone looking at it (and me later when i forget what it does) would know how it works and what does what. The Shop got chests added to it which can be opened with keys that you can also now collect in the dungeons. Thats pretty much it I did, im probably forgetting a few things but oh well. Im planing on making my first ship soon so Im trying to get it working well enough for people to use it"}, {"post_id": 10037, "ts": "2026-06-16T16:27:45Z", "hours": 5.34, "text": "I redid the enemy spawning yet again, but itâs much simpler and easier to modify now. Iâve also been working on enemy movement using the threading library to run multiple loops or functions at a time. Iâve also been working on trying to make the generation for each dungeon level better. One of my friends gave me the idea of having some rooms or something in each level, however there is a few bugs with it, in the photo below you can see it generates a path that just goes straight across the map whenever a room spawns on the initial path, and players can crash the game if they try to leave the map, so I need to fix both of those"}, {"post_id": 5402, "ts": "2026-06-09T04:50:03Z", "hours": 4.0883, "text": "I decided to redo some of the enemy/gold generation which toke me a while since I had to change it from doing a random percent of each row to a random percent of the whole map. I also worked on making some UIs (idk if thatâs what these are technically called) for choosing to enter a dungeon, open the shop etc. Ive also started working on switching the way the enemies work."}, {"post_id": 3892, "ts": "2026-06-07T00:00:24Z", "hours": 5.6181, "text": "I recolored some of the map and i added enemyâs and gold. The gold you can gather and get 1-3. The enemyâs take a life from you, if you lose all 5 lives you lose. I fixed a few visual bugs and i split the file into 2 files so i can use import and the call function to be able to do multiple levels. I also scaled the map up more so its easier to see."}, {"post_id": 3223, "ts": "2026-06-06T01:43:38Z", "hours": 2.2025, "text": "I mostly redid the random map generation making the user able to actually reach the end of the dungeon without getting stuck. I also added some color to it so the walls and floor are easy to see. I also added a safety check on the player movement and map generation so the player couldnât move outside of the map. I started to make it more interactive."}, {"post_id": 2492, "ts": "2026-06-05T00:43:04Z", "hours": 2.2664, "text": "I worked on setting up a random map generator and making it so the player could move around the map. I then added a fix that prevents the players from going through the walls. I then also setup a basic treasure/gold gathering system. After a bit I decided to redo most of the map generation and Iâm still working on it. I didnât realize I was supposed to have photos so this one is of a older version of it from my phone"}]...
|
Title: Dungeon-Adventure by @Reid | Stardance
Dev Title: Dungeon-Adventure by @Reid | Stardance
Devlogs:
I worked on setting up a random map generator and making it so the player could move around the map. I then added a fix that prevents the players from going through the walls. I then also setup a basic treasure/gold gathering system. After a bit I decided to redo most of the map generation and Iâm still working on it. I didnât realize I was supposed to have photos so this one is of a older version of it from my phone
I mostly redid the random map generation making the user able to actually reach the end of the dungeon without getting stuck. I also added some color to it so the walls and floor are easy to see. I also added a safety check on the player movement and map generation so the player couldnât move outside of the map. I started to make it more interactive.
I recolored some of the map and i added enemyâs and gold. The gold you can gather and get 1-3. The enemyâs take a life from you, if you lose all 5 lives you lose. I fixed a few visual bugs and i split the file into 2 files so i can use import and the call function to be able to do multiple levels. I also scaled the map up more so its easier to see.
I decided to redo some of the enemy/gold generation which toke me a while since I had to change it from doing a random percent of each row to a random percent of the whole map. I also worked on making some UIs (idk if thatâs what these are technically called) for choosing to enter a dungeon, open the shop etc. Ive also started working on switching the way the enemies work.
I redid the enemy spawning yet again, but itâs much simpler and easier to modify now. Iâve also been working on enemy movement using the threading library to run multiple loops or functions at a time. Iâve also been working on trying to make the generation for each dungeon level better. One of my friends gave me the idea of having some rooms or something in each level, however there is a few bugs with it, in the photo below you can see it generates a path that just goes straight across the map whenever a room spawns on the initial path, and players can crash the game if they try to leave the map, so I need to fix both of those
This is probably my longest work period between devlogs, I did also take like a week break in between them to. Im also hoping i can ship the project after this I worked on making the enemy movement smoother and work better, i decided to not use threading to run multiple loops since i didnât like how it worked so i just did it semi turn based where the player moves and the enemies will then move, but its so quick the player can just hold down a button to move and it will move them multiple spaces seamlessly. I also worked on polishing the UIs like the Inventory and Shop so that they look better, they now have color and look nice. Something else i worked on was making a Info option too that tells you a little about the game I worked on making the code look cleaner and I added a bunch of comments so that anyone looking at it (and me later when i forget what it does) would know how it works and what does what. The Shop got chests added to it which can be opened with keys that you can also now collect in the dungeons. Thats pretty much it I did, im probably forgetting a few things but oh well. Im planing on making my first ship soon so Im trying to get it working well enough for people to use it
So I took a break and itâs been a while since I worked on the project but I did finally start working on it again. Iâve decided that since itâs only compatible with some Linux devices I would make a html version that can run in browsers, I did some research and found out I could use pyodide to run Python in a html/javascript environment so I started using that and basically copy pasted most my code into the new file, I have had to modify some of the code like the movement and coloration to work with JavaScript instead. I also decided to make going into the different menus have buttons instead of pushing keys, but Iâm having a lot of trouble with getting the buttons to display at the right time and function correctly. Hopefully I can get them working right by the next devlog. (P.S. Iâm new to html and JavaScript so I donât fully know how it works and I kinda feel like buttons should be easier and Iâm just missing something, but maybe not and they really are just hard.)
Ships:
This is my first ship so I hope you all enjoy it. I made a Dungeon Adventure game that you run in your terminal or shell depending on what device you use. You can go through different dungeon levels that are randomly generated each time you enter a new one. You can also buy lifes or open chests in the shop using gold or keys that you can gather from withen the dungeon, but be careful because there are enemies that can attack you as well so keep an eye on your life and where the enemies are. Some challenges I faced while making this project were figuring out how to make the enemy movement and make it work well, I would say it was probably one of my most difficult and common problems because i had to rewrite how the enemies spawned and functioned over and over until i figured it out. I also had problems with making the small room generations withen the dungeon, because they would constantly keep generating paths that went outside the map causing a crash and i couldnt figure out why for the longest time (like most times with a problem like this it was a kinda simple solution but i still took forever to figure it out). I think my favorite part was making the random map generation because I found it fun to make and I built an algorithm to do it basically from the ground up which was exciting. I did have alot of problems with this too but I figured them out as well. This game is run in a terminal and you do need external libraries such as readchar and colorama to run the program. It comes in 2 files one is the main one you run and the other has a bunch of functions that are called from the first one or even from other functions because alot of the things need to be repeated constantly. Anyways i hope you guys enjoy this project and that it works for you (because im kinda scared that I uploaded something wrong or something to where it wont work for you guys) So have fun and let me know if anything isnt working right by commenting on one of my devlogs or contacting me in Slack (Jason Challand)...
|
1793271181.6739633
|
1793357581.6739633
|
|
7471
|
Untitled project by @sushidagoat185 | Stardance
|
sushidagoat185
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Untitled project by @sushidagoat185 | Stard Title: Untitled project by @sushidagoat185 | Stardance...
|
1792749207.9026606
|
1793354007.9026606
|
|
7472
|
rumi by @subhansh | Stardance
|
|
1
|
106.0
|
0
|
[{"ship_num": 1, "ts": "2026-0 [{"ship_num": 1, "ts": "2026-07-10T07:38:59Z", "hours": 106.0, "multiplier": null, "payout": 0, "blessing": "neutral", "text": "rumi - research unified machine intelligence⌠rumi is an actual autonomous AI scientist designed to discover, not just describe. Powered by a 22 phase, 2 tracks scientific discovery pipeline, RUMI analyzes research literature, builds knowledge representations, identifies gaps in existing explanations, generates novel hypotheses, constructs mathematical and causal models, produces falsifiable predictions, and evaluates competing theories through automated scientific review. Operating across 17 domains, RUMI is built as a general-purpose discovery engine whose objective is to accelerate human research by systematically exploring explanations that may not yet exist in the literature while maintaining transparency, critique, and testability throughout the discovery process. NOTE :- rumi still cant do end to end novel discoveries on her own but she can provide novel ideas with grounded mechanisms and derivations and stronger the reasoning of a model = better results from rumi.. that said⌠once i get my hands on a frontier llm ill plug it into rumi and develop her more Proof that rumi works :- rumi has helped me publish 1 peer reviewed research paper on analytical number theory (DOI :- 10.5281/zenodo.21107958)", "repo": "https://github.com/subhansh-dev/rumi", "demo": "https://rumi-nu.vercel.app"}]...
|
[{"post_id": 11764, "ts": "202 [{"post_id": 11764, "ts": "2026-06-19T04:39:49Z", "hours": 16.0183, "text": "rumi-Devlog #9 đđşâ today was mostly spent making RUMI faster because some runs were taking an absolutely stupid amount of time đ provider stack changes spent a while profiling the provider chain and found a few cursed bottlenecks. disabled Fireworks (credits died, every failed request was wasting ~90s) switched GLM routing to Kimi K2.6 via a second NVIDIA key tested Cerebras-first vs NVIDIA-first routing results were funny: Cerebras = ~2s/call but discoveries averaged around C grade NVIDIA = ~65s/call but much stronger reports so now different parts of the pipeline use different providers depending on the task. phase 6.5 speedup equation generation was committing crimes. rewrote the whole thing to run in parallel (6 workers). went from: ~16 min to ~1 min for a typical 15 mechanism loop. also finally fixed the legendary: 0/15 equations generated bug đşđ literature search all literature sources now fetch concurrently: arXiv PubMed CrossRef OpenAlex etc runtime dropped from ~2 mins to ~64 seconds. RUMI TUI probably my favourite thing from today. built a proper Textual-based TUI so i can actually see what RUMI is doing instead of staring at a terminal for 10 hours. added: real-time phase tracking live agent status brain module dashboard discovery controls /brain /status /dream /memory /hypothesis also fixed a bunch of broken brain module APIs that were pretending to work đ random bug of the day RUMI was occasionally turning english into ancient cursed language and generating stuff like: withinamy instead of actual words. fixed topic truncation so it now cuts at word boundaries. current state provider stack healthier equation generation WAY faster literature search faster TUI finally usable brain modules actually verified at boot still spending most of my life waiting for discovery runs to finish though đĽ"}, {"post_id": 9228, "ts": "2026-06-15T14:33:23Z", "hours": 23.0847, "text": "rumi-Devlog #8 đ Current status: rumis smarter, my codebase survived, and my sanity is questionable đ A huge part of this session wasnât actually spent improving discovery quality. It was spent improving everything around RUMI. As RUMIâs discovery pipeline keeps getting bigger, debugging has become a nightmare. Reading raw JSON reports, searching through logs, and manually inspecting hundreds of entities was getting painful. So I built a full visualization and monitoring system for RUMI. The dashboard can now visualize: discovery reports theory competitions adversarial testing results predictions refinement stages scoring breakdowns knowledge graphs paper collections hypothesis memory all from a single interface. Knowledge Graph Visualization One thing Iâve wanted for a while was being able to actually SEE what RUMI was discovering. Not read it. See it. So I built an interactive graph explorer using vis.js. Every entity type gets its own visual identity. Nodes glow based on type. Edges scale with confidence. Hovering shows metadata, paper counts, and relationships. The graph uses ForceAtlas physics and overlap avoidance to make large discovery graphs easier to navigate. Watching hundreds of concepts connect together is honestly way more useful than reading raw graph dumps. đĽ Discovery Report Explorer RUMI reports have become massive. Some are hundreds of kilobytes. Some are over a thousand lines. Reading those manually sucks. So I built dedicated views for: theory competition adversarial evaluation predictions scoring derivations peer review refinement stages which makes analyzing runs significantly easier. Provider Stack Surgery đ While testing the new recurrent architecture I discovered a huge bottleneck. Xiaomi MiMo. It was slow. Constantly rate limiting. Burning through provider chains. And starving downstream phases. This turned out to be the reason molecule generation kept randomly failing. So MiMo got removed from the routing stack. After testing 17 API keys across all providers: 13 healthy 4 problematic Current routing stack includes: NVIDIA DeepSeek Kimi K2.6 Cerebras GPT-OSS-120B Groq Gemini Fireworks Still chasing down some timeout and rate limit issues but overall the stack is way healthier now. Discovery Testing Used the upgraded architecture to stress test RUMI on several difficult domains. One of the most interesting runs was Molecular Glue Drug Discovery. Current run stats: 16 hidden variables 15 mechanisms per loop 7 predictions multiple adversarial survivors recurrent refinement loops enabled The architecture is definitely producing stronger outputs than before. meow meow meow"}, {"post_id": 7717, "ts": "2026-06-12T21:37:49Z", "hours": 19.0981, "text": "rumi-Devlog #7 đ This was probably one of the most frustrating and important RUMI development sessions so far. Most people only see the final discovery reports. They donât see the chaos that happens before RUMI can generate a single theory. And trust me, there was a lot of chaos. đ Bug Fixes, Bug Fixes, More Bug Fixes I started this session thinking Iâd make a few improvements and continue testing. That did not happen. RUMIâs architecture has reached a point where changing one thing breaks three others. Every time I fixed 3 bugs, another 5 appeared from somewhere else. At one point I genuinely crashed out because every run ended with a new issue that didnât even exist before. It felt like: fix_bug() spawn_more_bugs(count=5) Turns out thereâs a reason people say: âIf itâs working, donât touch it.â Unfortunately I touched it. A lot. đĽ The Problem Even though RUMI was producing interesting discoveries, I still wasnât satisfied. The reports were often interesting, creative, and sometimes novel, but they still felt too generic. The pipeline would generate a theory, evaluate it once, pick a winner, and move on. I wanted RUMI to spend more time refining ideas instead of treating every discovery as a one-shot attempt. Falling Into The OpenMythos Rabbit Hole While researching I came across OpenMythos, a community attempt to reverse engineer Anthropicâs rumored Mythos architecture. Nobody actually knows how Mythos works, but the community has proposed several reasoning patterns that might explain Claude-style outputs. I spent hours reading through it asking myself: âWhat parts of this could actually work inside RUMI?â Not copying. Adapting. The Biggest Change Iâve Ever Made To RUMI đĽ One idea stood out immediately: recurrent reasoning loops. Before: Mechanisms â Predictions â Theory Competition â Winner One pass. One shot. If mechanism generation was weak, everything downstream became weak too. So I rebuilt a huge section of the pipeline. Now RUMI performs a 3-stage recurrent refinement process: Loop 1 â Exploration Maximum creativity Maximum diversity Broad theory generation Loop 2 â Refinement Survivors re-enter the pipeline Focus on evidence, consistency, and mechanism quality Loop 3 â Convergence Lowest creativity Highest rigor Final selection of the strongest explanation Instead of living or dying from a single generation pass, theories now get multiple opportunities to evolve. Preventing Theory Drift Another idea I adapted was evidence grounding injection. Recursive systems often suffer from theory drift, where ideas slowly become disconnected from the evidence that originally generated them. To prevent this, every refinement loop continuously receives: Literature evidence Contradictions Knowledge gaps Observed anomalies This keeps discoveries grounded instead of drifting into fantasy. Smarter Computation I also added convergence-aware halting. Not every scientific problem deserves the same amount of computation. Some topics converge quickly while others require deeper exploration. RUMI can now monitor improvements between loops and eventually stop refining once discoveries begin stabilizing. lwk so much happened in these last 17hrs that i cant even include everything in here cause of the word limit đđĽ"}, {"post_id": 6459, "ts": "2026-06-10T21:25:00Z", "hours": 15.2386, "text": "rumi - Devlog #6 June 11, 2026 so yea⌠pulled an all nighter for this one đĽ most of the work this time wasnât adding entirely new systems. it was stress testing RUMI hard enough to find where the architecture starts breaking. and honestly i found a lot more than i expected. Track B Architecture Improvements spent a lot of time refining Track B after the initial curiosity pipeline integration. instead of just generating curiosity questions, it now runs through much more of the actual discovery stack. lots of internal changes here that donât really show up visually but massively affect the quality of the final reports. đ Constraint Pipeline Fix found a pretty nasty issue where parts of the generated curiosity constraints werenât making it all the way through the pipeline. after fixing it, the results became immediately obvious. latest dual-track run: Track A Unique Theories: 9 Track B Unique Theories: 8 Shared Theories: 0 which is honestly one of the strongest signals iâve seen so far that the dual-track system is actually working as intended. both tracks are now exploring completely different hypothesis spaces instead of converging on the same ideas. Claude Fable 5 Experiment managed to get temporary access to Claude Fable 5 and immediately decided to throw RUMI at it. before i could even test it though⌠the integration broke đ spent a while fixing provider compatibility issues and getting everything working again. eventually got a discovery run started on: What happens to information when it crosses a black hole event horizon? Fable managed to reach roughly Phase 8.5 before the free trial looked at RUMIâs request count and basically said: aight imma head out đ What Fable Revealed this ended up being way more valuable than i expected. while comparing Fableâs outputs with the models i normally run, i noticed a massive difference. most models tend to generate: Hidden Variable â Mechanism Description â Prediction Fable was generating: Hidden Variable â Mechanism â Equation â Parameter Extraction â Derivation â Numerical Validation â Prediction actual equations. actual variables. actual derivation chains. actual numerical checks. đĽ Switching Back To MiMo after the Fable credits died, i moved everything back to MiMo and started comparing outputs. and thatâs when another weakness became obvious. the pipeline itself wasnât failing. MiMo simply wasnât producing the mathematical depth needed for the mechanism stage. which means the current bottleneck isnât discovery anymore. itâs mathematical formalization. Mathematical Formalization Architecture one of the most interesting things i noticed while testing Claude Fable 5 was how differently it handled mechanism generation. most models would generate something like: Hidden Variable â Mechanism Description â Prediction while Fable was generating: Hidden Variable â Mechanism â Equation â Parameter Extraction â Derivation â Numerical Validation â Prediction and honestly that immediately exposed one of RUMIâs biggest weaknesses. theories and mechanisms were already being generated. the math wasnât. so instead of just complaining about it, i started redesigning the mechanism pipeline around that pattern. RUMI now has the foundations for a much more quantitative discovery process where mechanisms arenât just descriptions anymore, which is lwk tuff ig đĽâ equations variables derivations parameter extraction numerical validation Switching Back To MiMo đĽ after Fable ran out of credits i moved everything back to MiMo and started testing the updated architecture and still doing it cause mimo is like hella slow bruh normally umi takes upto 30-40 mins but now with mimo shes taking aorund 2hr 30mins đ"}, {"post_id": 5243, "ts": "2026-06-08T23:06:33Z", "hours": 10.65, "text": "Devlog #5 â RUMI Dual-Track Architecture so as yall know yesterday i finished adding the curiosity pipeline (wait⌠you dont know about this? naaađgo read yesterdayâs devlog >:3âŚWUT ? GO RN) but yea that was mostly just the foundation today was about actually refining it and wiring it into the rest of RUMI basically i ended up bringing a lot of the Track A discovery pipeline into Track B too because like⌠why wouldnt i đ not just straight up copy-pasting it though because then both tracks would end up doing the same thing instead i reused parts of the architecture while keeping the curiosity systems separate the biggest addition was making the generated curiosity constraints actually flow into the discovery process itself which sounds simple until you spend half the day finding out they werent actually being propagated through the pipeline properly đ and then came the debugging arc đ DNS resolution failures causing discovery runs to randomly die đž (like wdym rumi requested for more papers and data and dns forgetting that internet exists) curiosity constraints not propagating into the pipelineđŤ Semantic Scholar 429 rate limits đĽ Gemini 503 errors đ˘ free-tier rate limits everywhere (she makes like 6-7k requests in one run )đşđĽ and somehow RUMI has become so LLM hungry that i had to run her on 13 API keys across 3 different providers just to keep discovery sessions alive đ (lwk running on hope and sum duct tape đĽđĽ) there was genuinely a point where i was about to crash out because providers kept rate limiting, APIs were dying, runs were hanging, and everything was fighting me at once đđĽđ but after all that RUMI finally completed a full dual-track discovery run topic: âWhy does the universe expand faster than expected?â results: Track A Winner â Scale-Dependent Effective Gravity Track B Winner â Temporal Vacuum Shear Framework and this is where things got interesting Track A and Track B finished with: 0 shared theories <â(đ) different winners different hidden variables different contradiction sets different mechanisms 0 runtime errors RUMI generated: 17 theories 30 mechanisms 22 hidden variables 12 contradictions 12 predictions from a single discovery session đ and honestly the most important number isnt the score its this: Shared Theories: 0 because it means Track B isnt just making a slightly remixed version of Track A anymore both tracks started from the same question, same literature, same knowledge base, and still ended up exploring completely different hypothesis spaces which is pretty much the entire reason i built the curiosity system in the first place đđĽ still more work to do, but today feels like one of the biggest architecture milestones for RUMI so farđđĽ lwk gonna try to improve GFlowNets tmrw hereâs the updated pipeline diagram :-"}, {"post_id": 4578, "ts": "2026-06-07T22:28:00Z", "hours": 9.3764, "text": "Stardance Devlog #3 June 7, 2026 Today I realized RUMIâs biggest limitation wasnât math, evidence, reviewers, or the 17-stage pipeline. It was curiosity. For days Iâve been making RUMI better at reading papers, finding gaps, generating hypotheses, scoring theories, and validating ideas, which actually improved her ALLAT⌠but still something felt missing. Then it hit me. RUMI could analyze research. RUMI could synthesize research. RUMI could critique research. But RUMI never stopped and asked: âWhy does this happen at all?â And thatâs where most discoveries beginđĽ So today I built Phase 0 â The Curiosity Engine. Before reading papers. Before building knowledge graphs. Before generating mechanisms. RUMI now starts by questioning the problem itself. What could be causing this? What assumptions are we making? What if the accepted explanation is incomplete? What are we not looking at? Basically, RUMI now wonders before she researchesđ The first implementation was a disaster. I merged curiosity directly into the main pipeline and immediately turned the architecture into spaghetti. So I rebuilt it. Now there are two separate paths: Conventional RUMI Reads Analyzes Verifies Curious RUMI Questions Explores Challenges assumptions Both run independently and merge into the final reportđ The funny part is that after building theory tournaments, Bayesian reasoning, evidence systems, mathematical verification, observability checks, and dozens of other modules⌠The biggest upgrade might have been teaching RUMI something humans learn as children: CuriosityđĽđ Todayâs Progress Phase 0 Curiosity Engine â
Dual-Pipeline Architecture â
Evidence Extraction Improvements â
Observability Checks â
Mechanism Completeness Scoring â
My Sanity â See you tomorrow. Hopefully before RUMI starts asking questions that create more problems than answers. đĽđ"}, {"post_id": 3851, "ts": "2026-06-06T22:22:37Z", "hours": 2.2931, "text": "spent basically the whole day inside RUMIâs discovery engine and ngl it was pain đ fixed 50+ bugs across the pipeline. some of the biggest ones were: empty hypotheses theory tournament crashing unicode errors on windows graph contamination novelty scoring issues mechanism validation deleting everything đ also built a few new systems: math verification engine counterfactual/âwhat ifâ hypothesis generation entity enrichment (NASA, PubChem, UniProt, PDB, etc.) better novelty scoring tournament winner override before the fixes, some discoveries were scoring: 0/100 40/100 45/100 after everything: FRB Magnetar: 66/100 (B) KRAS G12D: 65/100 (B) the coolest run today was on Fast Radio Bursts and magnetars. RUMI went through 69 papers, generated new hidden variables, built mechanisms around them, made testable predictions, then tried to destroy its own ideas through adversarial testing. finally it actually felt like the pipeline was working end-to-end instead of randomly exploding lol đ still got a long way to go though. biggest weaknesses right now are novelty and mathematical rigor. RUMI is getting better at explaining science, now I need to make her better at finding genuinely new stuff. back to debugging tomorrow frfr đđĽđ"}, {"post_id": 3528, "ts": "2026-06-06T14:42:38Z", "hours": 2.1419, "text": "yayaya finallyy finally the basic tui is working and the frontend-backend integration is almost there"}, {"post_id": 2865, "ts": "2026-06-05T18:48:30Z", "hours": 7.7036, "text": "12hr 726mins in and still not finished debugging đđĽ"}]...
|
Title: rumi by @subhansh | Stardance
Devlogs:
12h Title: rumi by @subhansh | Stardance
Devlogs:
12hr 726mins in and still not finished debugging đđĽ
yayaya finallyy finally the basic tui is working and the frontend-backend integration is almost there
spent basically the whole day inside RUMIâs discovery engine and ngl it was pain đ fixed 50+ bugs across the pipeline. some of the biggest ones were: empty hypotheses theory tournament crashing unicode errors on windows graph contamination novelty scoring issues mechanism validation deleting everything đ also built a few new systems: math verification engine counterfactual/âwhat ifâ hypothesis generation entity enrichment (NASA, PubChem, UniProt, PDB, etc.) better novelty scoring tournament winner override before the fixes, some discoveries were scoring: 0/100 40/100 45/100 after everything: FRB Magnetar: 66/100 (B) KRAS G12D: 65/100 (B) the coolest run today was on Fast Radio Bursts and magnetars. RUMI went through 69 papers, generated new hidden variables, built mechanisms around them, made testable predictions, then tried to destroy its own ideas through adversarial testing. finally it actually felt like the pipeline was working end-to-end instead of randomly exploding lol đ still got a long way to go though. biggest weaknesses right now are novelty and mathematical rigor. RUMI is getting better at explaining science, now I need to make her better at finding genuinely new stuff. back to debugging tomorrow frfr đđĽđ
Stardance Devlog #3 June 7, 2026 Today I realized RUMIâs biggest limitation wasnât math, evidence, reviewers, or the 17-stage pipeline. It was curiosity. For days Iâve been making RUMI better at reading papers, finding gaps, generating hypotheses, scoring theories, and validating ideas, which actually improved her ALLAT⌠but still something felt missing. Then it hit me. RUMI could analyze research. RUMI could synthesize research. RUMI could critique research. But RUMI never stopped and asked: âWhy does this happen at all?â And thatâs where most discoveries beginđĽ So today I built Phase 0 â The Curiosity Engine. Before reading papers. Before building knowledge graphs. Before generating mechanisms. RUMI now starts by questioning the problem itself. What could be causing this? What assumptions are we making? What if the accepted explanation is incomplete? What are we not looking at? Basically, RUMI now wonders before she researchesđ The first implementation was a disaster. I merged curiosity directly into the main pipeline and immediately turned the architecture into spaghetti. So I rebuilt it. Now there are two separate paths: Conventional RUMI Reads Analyzes Verifies Curious RUMI Questions Explores Challenges assumptions Both run independently and merge into the final reportđ The funny part is that after building theory tournaments, Bayesian reasoning, evidence systems, mathematical verification, observability checks, and dozens of other modules⌠The biggest upgrade might have been teaching RUMI something humans learn as children: CuriosityđĽđ Todayâs Progress Phase 0 Curiosity Engine â
Dual-Pipeline Architecture â
Evidence Extraction Improvements â
Observability Checks â
Mechanism Completeness Scoring â
My Sanity â See you tomorrow. Hopefully before RUMI starts asking questions that create more problems than answers. đĽđ
Devlog #5 â RUMI Dual-Track Architecture so as yall know yesterday i finished adding the curiosity pipeline (wait⌠you dont know about this? naaađgo read yesterdayâs devlog >:3âŚWUT ? GO RN) but yea that was mostly just the foundation today was about actually refining it and wiring it into the rest of RUMI basically i ended up bringing a lot of the Track A discovery pipeline into Track B too because like⌠why wouldnt i đ not just straight up copy-pasting it though because then both tracks would end up doing the same thing instead i reused parts of the architecture while keeping the curiosity systems separate the biggest addition was making the generated curiosity constraints actually flow into the discovery process itself which sounds simple until you spend half the day finding out they werent actually being propagated through the pipeline properly đ and then came the debugging arc đ DNS resolution failures causing discovery runs to randomly die đž (like wdym rumi requested for more papers and data and dns forgetting that internet exists) curiosity constraints not propagating into the pipelineđŤ Semantic Scholar 429 rate limits đĽ Gemini 503 errors đ˘ free-tier rate limits everywhere (she makes like 6-7k requests in one run )đşđĽ and somehow RUMI has become so LLM hungry that i had to run her on 13 API keys across 3 different providers just to keep discovery sessions alive đ (lwk running on hope and sum duct tape đĽđĽ) there was genuinely a point where i was about to crash out because providers kept rate limiting, APIs were dying, runs were hanging, and everything was fighting me at once đđĽđ but after all that RUMI finally completed a full dual-track discovery run topic: âWhy does the universe expand faster than expected?â results: Track A Winner â Scale-Dependent Effective Gravity Track B Winner â Temporal Vacuum Shear Framework and this is where things got interesting Track A and Track B finished with: 0 shared theories <â(đ) different winners different hidden variables different contradiction sets different mechanisms 0 runtime errors RUMI generated: 17 theories 30 mechanisms 22 hidden variables 12 contradictions 12 predictions from a single discovery session đ and honestly the most important number isnt the score its this: Shared Theories: 0 because it means Track B isnt just making a slightly remixed version of Track A anymore both tracks started from the same question, same literature, same knowledge base, and still ended up exploring completely different hypothesis spaces which is pretty much the entire reason i built the curiosity system in the first place đđĽ still more work to do, but today feels like one of the biggest architecture milestones for RUMI so farđđĽ lwk gonna try to improve GFlowNets tmrw hereâs the updated pipeline diagram :-
rumi - Devlog #6 June 11, 2026 so yea⌠pulled an all nighter for this one đĽ most of the work this time wasnât adding entirely new systems. it was stress testing RUMI hard enough to find where the architecture starts breaking. and honestly i found a lot more than i expected. Track B Architecture Improvements spent a lot of time refining Track B after the initial curiosity pipeline integration. instead of just generating curiosity questions, it now runs through much more of the actual discovery stack. lots of internal changes here that donât really show up visually but massively affect the quality of the final reports. đ Constraint Pipeline Fix found a pretty nasty issue where parts of the generated curiosity constraints werenât making it all the way through the pipeline. after fixing it, the results became immediately obvious. latest dual-track run: Track A Unique Theories: 9 Track B Unique Theories: 8 Shared Theories: 0 which is honestly one of the strongest signals iâve seen so far that the dual-track system is actually working as intended. both tracks are now exploring completely different hypothesis spaces instead of converging on the same ideas. Claude Fable 5 Experiment managed to get temporary access to Claude Fable 5 and immediately decided to throw RUMI at it. before i could even test it though⌠the integration broke đ spent a while fixing provider compatibility issues and getting everything working again. eventually got a discovery run started on: What happens to information when it crosses a black hole event horizon? Fable managed to reach roughly Phase 8.5 before the free trial looked at RUMIâs request count and basically said: aight imma head out đ What Fable Revealed this ended up being way more valuable than i expected. while comparing Fableâs outputs with the models i normally run, i noticed a massive difference. most models tend to generate: Hidden Variable â Mechanism Description â Prediction Fable was generating: Hidden Variable â Mechanism â Equation â Parameter Extraction â Derivation â Numerical Validation â Prediction actual equations. actual variables. actual derivation chains. actual numerical checks. đĽ Switching Back To MiMo after the Fable credits died, i moved everything back to MiMo and started comparing outputs. and thatâs when another weakness became obvious. the pipeline itself wasnât failing. MiMo simply wasnât producing the mathematical depth needed for the mechanism stage. which means the current bottleneck isnât discovery anymore. itâs mathematical formalization. Mathematical Formalization Architecture one of the most interesting things i noticed while testing Claude Fable 5 was how differently it handled mechanism generation. most models would generate something like: Hidden Variable â Mechanism Description â Prediction while Fable was generating: Hidden Variable â Mechanism â Equation â Parameter Extraction â Derivation â Numerical Validation â Prediction and honestly that immediately exposed one of RUMIâs biggest weaknesses. theories and mechanisms were already being generated. the math wasnât. so instead of just complaining about it, i started redesigning the mechanism pipeline around that pattern. RUMI now has the foundations for a much more quantitative discovery process where mechanisms arenât just descriptions anymore, which is lwk tuff ig đĽâ equations variables derivations parameter extraction numerical validation Switching Back To MiMo đĽ after Fable ran out of credits i moved everything back to MiMo and started testing the updated architecture and still doing it cause mimo is like hella slow bruh normally umi takes upto 30-40 mins but now with mimo shes taking aorund 2hr 30mins đ
rumi-Devlog #7 đ This was probably one of the most frustrating and important RUMI development sessions so far. Most people only see the final discovery reports. They donât see the chaos that happens before RUMI can generate a single theory. And trust me, there was a lot of chaos. đ Bug Fixes, Bug Fixes, More Bug Fixes I started this session thinking Iâd make a few improvements and continue testing. That did not happen. RUMIâs architecture has reached a point where changing one thing breaks three others. Every time I fixed 3 bugs, another 5 appeared from somewhere else. At one point I genuinely crashed out because every run ended with a new issue that didnât even exist before. It felt like: fix_bug() spawn_more_bugs(count=5) Turns out thereâs a reason people say: âIf itâs working, donât touch it.â Unfortunately I touched it. A lot. đĽ The Problem Even though RUMI was producing interesting discoveries, I still wasnât satisfied. The reports were often interesting, creative, and sometimes novel, but they still felt too generic. The pipeline would generate a theory, evaluate it once, pick a winner, and move on. I wanted RUMI to spend more time refining ideas instead of treating every discovery as a one-shot attempt. Falling Into The OpenMythos Rabbit Hole While researching I came across OpenMythos, a community attempt to reverse engineer Anthropicâs rumored Mythos architecture. Nobody actually knows how Mythos works, but the community has proposed several reasoning patterns that might explain Claude-style outputs. I spent hours reading through it asking myself: âWhat parts of this could actually work inside RUMI?â Not copying. Adapting. The Biggest Change Iâve Ever Made To RUMI đĽ One idea stood out immediately: recurrent reasoning loops. Before: Mechanisms â Predictions â Theory Competition â Winner One pass. One shot. If mechanism generation was weak, everything downstream became weak too. So I rebuilt a huge section of the pipeline. Now RUMI performs a 3-stage recurrent refinement process: Loop 1 â Exploration Maximum creativity Maximum diversity Broad theory generation Loop 2 â Refinement Survivors re-enter the pipeline Focus on evidence, consistency, and mechanism quality Loop 3 â Convergence Lowest creativity Highest rigor Final selection of the strongest explanation Instead of living or dying from a single generation pass, theories now get multiple opportunities to evolve. Preventing Theory Drift Another idea I adapted was evidence grounding injection. Recursive systems often suffer from theory drift, where ideas slowly become disconnected from the evidence that originally generated them. To prevent this, every refinement loop continuously receives: Literature evidence Contradictions Knowledge gaps Observed anomalies This keeps discoveries grounded instead of drifting into fantasy. Smarter Computation I also added convergence-aware halting. Not every scientific problem deserves the same amount of computation. Some topics converge quickly while others require deeper exploration. RUMI can now monitor improvements between loops and eventually stop refining once discoveries begin stabilizing. lwk so much happened in these last 17hrs that i cant even include everything in here cause of the word limit đđĽ
rumi-Devlog #8 đ Current status: rumis smarter, my codebase survived, and my sanity is questionable đ A huge part of this session wasnât actually spent improving discovery quality. It was spent improving everything around RUMI. As RUMIâs discovery pipeline keeps getting bigger, debugging has become a nightmare. Reading raw JSON reports, searching through logs, and manually inspecting hundreds of entities was getting painful. So I built a full visualization and monitoring system for RUMI. The dashboard can now visualize: discovery reports theory competitions adversarial testing results predictions refinement stages scoring breakdowns knowledge graphs paper collections hypothesis memory all from a single interface. Knowledge Graph Visualization One thing Iâve wanted for a while was being able to actually SEE what RUMI was discovering. Not read it. See it. So I built an interactive graph explorer using vis.js. Every entity type gets its own visual identity. Nodes glow based on type. Edges scale with confidence. Hovering shows metadata, paper counts, and relationships. The graph uses ForceAtlas physics and overlap avoidance to make large discovery graphs easier to navigate. Watching hundreds of concepts connect together is honestly way more useful than reading raw graph dumps. đĽ Discovery Report Explorer RUMI reports have become massive. Some are hundreds of kilobytes. Some are over a thousand lines. Reading those manually sucks. So I built dedicated views for: theory competition adversarial evaluation predictions scoring derivations peer review refinement stages which makes analyzing runs significantly easier. Provider Stack Surgery đ While testing the new recurrent architecture I discovered a huge bottleneck. Xiaomi MiMo. It was slow. Constantly rate limiting. Burning through provider chains. And starving downstream phases. This turned out to be the reason molecule generation kept randomly failing. So MiMo got removed from the routing stack. After testing 17 API keys across all providers: 13 healthy 4 problematic Current routing stack includes: NVIDIA DeepSeek Kimi K2.6 Cerebras GPT-OSS-120B Groq Gemini Fireworks Still chasing down some timeout and rate limit issues but overall the stack is way healthier now. Discovery Testing Used the upgraded architecture to stress test RUMI on several difficult domains. One of the most interesting runs was Molecular Glue Drug Discovery. Current run stats: 16 hidden variables 15 mechanisms per loop 7 predictions multiple adversarial survivors recurrent refinement loops enabled The architecture is definitely producing stronger outputs than before. meow meow meow
rumi-Devlog #9 đđşâ today was mostly spent making RUMI faster because some runs were taking an absolutely stupid amount of time đ provider stack changes spent a while profiling the provider chain and found a few cursed bottlenecks. disabled Fireworks (credits died, every failed request was wasting ~90s) switched GLM routing to Kimi K2.6 via a second NVIDIA key tested Cerebras-first vs NVIDIA-first routing results were funny: Cerebras = ~2s/call but discoveries averaged around C grade NVIDIA = ~65s/call but much stronger reports so now different parts of the pipeline use different providers depending on the task. phase 6.5 speedup equation generation was committing crimes. rewrote the whole thing to run in parallel (6 workers). went from: ~16 min to ~1 min for a typical 15 mechanism loop. also finally fixed the legendary: 0/15 equations generated bug đşđ literature search all literature sources now fetch concurrently: arXiv PubMed CrossRef OpenAlex etc runtime dropped from ~2 mins to ~64 seconds. RUMI TUI probably my favourite thing from today. built a proper Textual-based TUI so i can actually see what RUMI is doing instead of staring at a terminal for 10 hours. added: real-time phase tracking live agent status brain module dashboard discovery controls /brain /status /dream /memory /hypothesis also fixed a bunch of broken brain module APIs that were pretending to work đ random bug of the day RUMI was occasionally turning english into ancient cursed language and generating stuff like: withinamy instead of actual words. fixed topic truncation so it now cuts at word boundaries. current state provider stack healthier equation generation WAY faster literature search faster TUI finally usable brain modules actually verified at boot still spending most of my life waiting for discovery runs to finish though đĽ
Ships:
rumi - research unified machine intelligence⌠rumi is an actual autonomous AI scientist designed to discover, not just describe. Powered by a 22 phase, 2 tracks scientific discovery pipeline, RUMI analyzes research literature, builds knowledge representations, identifies gaps in existing explanations, generates novel hypotheses, constructs mathematical and causal models, produces falsifiable predictions, and evaluates competing theories through automated scientific review. Operating across 17 domains, RUMI is built as a general-purpose discovery engine whose objective is to accelerate human research by systematically exploring explanations that may not yet exist in the literature while maintaining transparency, critique, and testability throughout the discovery process. NOTE :- rumi still cant do end to end novel discoveries on her own but she can provide novel ideas with grounded mechanisms and derivations and stronger the reasoning of a model = better results from rumi.. that said⌠once i get my hands on a frontier llm ill plug it into rumi and develop her more Proof that rumi works :- rumi has helped me publish 1 peer reviewed research paper on analytical number theory (DOI :- 10.5281/zenodo.21107958)...
|
1793271008.875876
|
1793357408.875876
|
|
7473
|
Untitled project by @fishjigs | Stardance
|
fishjigs
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Untitled project by @fishjigs | Stardance
|
1793181272.3260639
|
1793786072.3260639
|
|
7474
|
AXIS by @Sem_Acou | Stardance
|
Sem_Acou
|
1
|
3.0
|
0
|
[{"ship_num": 1, "ts": "2026-1 [{"ship_num": 1, "ts": "2026-10-27T20:46:12Z", "hours": 3.0, "multiplier": null, "payout": 0, "blessing": "neutral", "text": "I finished my balancin robot assembly! Today I finally finished my dual wheeled balancing robot that I spent such a long time making (mainly because of deliverytimes)! I did have to resolder my entire pcb because I soldered my motor driver on⌠upside down đŤŁBut all good now! I then integrated everything into the 3d printed frame. And voila, the robot is done!! ⌠for the hardware side Luckly I did make the code beforehand while waiting for deliveries and such! I also added a feature to steer the robot with my hands!!! (see video in the github âgalleryâ folder) I was very lucky that this code almost worked from the beginning, I only still need to tune my PID a bit more but since Stardance is ending I needed to ship nowđĽ I want to thank you all for following my journey here!!", "repo": "https://github.com/MrQuaq/AXIS", "demo": ""}]...
|
[{"post_id": 78859, "ts": "202 [{"post_id": 78859, "ts": "2026-10-11T19:51:12Z", "hours": 3.2489, "text": "My parts arrived! Hi again! Itâs been a long long time since my last devlog. The reason⌠MY PARTSSS! They finally arrived, in the mean time I printed the entire frame and prepared everything. The great news is! I already soldered everything, it went so smoothly. Only things left are testing if my pcb works aandd integrating it into the frame! See you (hopefully in not that long of a time)"}, {"post_id": 35803, "ts": "2026-07-27T13:56:11Z", "hours": 5.0336, "text": "I finally finished the CAD! Today I finally finished the 3D CAD model of my balancing robot called AXIS! I had some setbacks and difficulties along the way, but thatâs normal! Eventually I found a solution for all of them! Thatâs exactly what engineering is about! Letâs get printing Oh god⌠this will take daysđĽ"}, {"post_id": 31427, "ts": "2026-07-19T19:46:29Z", "hours": 6.2958, "text": "Itâs been a while! In the mean time I have been working on the frame for the robot! I mainly did some prototyping for the motor mounts. Which went⌠letâs say âcould have gone betterâđ
I had to print multiple protorypes because the motors have a fairly irregular shape, and Iâm not good enough to do it first tryâŚđ
Also, because of some mistakes when saving, there sometimes missed parts on the print that I did actually change. Which caused some errors in my brain. Next up is the PCB and battery housing (the head of the robot)"}, {"post_id": 16966, "ts": "2026-06-27T15:23:54Z", "hours": 4.5006, "text": "I spent way too long on this issue 𼲠You know, a project isnt a project when theres not one moment where you think that is SO over! That was today! I began testing some parts of the hardware before ordering the PCB and without putting it inside the frame. At first everything went smoothly (supicously smoothly), but then⌠My motors began doing some weird ahh stuff where they reacted way too late after I tilted the sensor. I thought, it must be some timing issue, or my code is just so poorly written that its running too slow. I tried some methods to make it faster (less lcd updates, etc.) Still nothing⌠Then I made a little debugging program to find out the frequency at which the code runs, and the delta time variable. Turns out that this wasnt the problem at all!! My frequency and delta time were perfect! ⌠Thats when I made the discovery that I didnt think about the PID values. I just set them to some number and didnât give it a second thought because I wont tune it anyways without the frame⌠But of course the motors wont respond as I hoped they would if the values werenât even close to what they need to beđ But weâre back babyyyyyy!!!! Next is the frame design and CAD!"}, {"post_id": 12725, "ts": "2026-06-20T19:59:42Z", "hours": 1.7178, "text": "Today I integrated the new IMU iâm using into the code. (also had some beef with the library) I edited my code to accomodate for the other IMU Iâm using (since i broke the previous one, whoooppss) After that I quickly tested the measured and filtered data, along with the pid output data. Next up Iâm gonna test my motors and encoders!"}, {"post_id": 10153, "ts": "2026-06-16T19:26:02Z", "hours": 0.4675, "text": "Have been working on this project for a while (before stardust). Iâm making a balancing robot using an IMU and some DC-motors with encoders. Iâm trying to make it controlable with an xbox controller and it uses a PID algorithm to balance itself. Right now I have: code for measuring, computing, controlling the motors and receiving inputs from the xbox controller(untested) a pcb design which i can mount breakout boards on Whatâs still up: testing the code a lot! getting the actual pcb 3d modelling the frame"}]...
|
Title: AXIS by @Sem_Acou | Stardance
Devlogs:
Hav Title: AXIS by @Sem_Acou | Stardance
Devlogs:
Have been working on this project for a while (before stardust). Iâm making a balancing robot using an IMU and some DC-motors with encoders. Iâm trying to make it controlable with an xbox controller and it uses a PID algorithm to balance itself. Right now I have: code for measuring, computing, controlling the motors and receiving inputs from the xbox controller(untested) a pcb design which i can mount breakout boards on Whatâs still up: testing the code a lot! getting the actual pcb 3d modelling the frame
Today I integrated the new IMU iâm using into the code. (also had some beef with the library) I edited my code to accomodate for the other IMU Iâm using (since i broke the previous one, whoooppss) After that I quickly tested the measured and filtered data, along with the pid output data. Next up Iâm gonna test my motors and encoders!
I spent way too long on this issue 𼲠You know, a project isnt a project when theres not one moment where you think that is SO over! That was today! I began testing some parts of the hardware before ordering the PCB and without putting it inside the frame. At first everything went smoothly (supicously smoothly), but then⌠My motors began doing some weird ahh stuff where they reacted way too late after I tilted the sensor. I thought, it must be some timing issue, or my code is just so poorly written that its running too slow. I tried some methods to make it faster (less lcd updates, etc.) Still nothing⌠Then I made a little debugging program to find out the frequency at which the code runs, and the delta time variable. Turns out that this wasnt the problem at all!! My frequency and delta time were perfect! ⌠Thats when I made the discovery that I didnt think about the PID values. I just set them to some number and didnât give it a second thought because I wont tune it anyways without the frame⌠But of course the motors wont respond as I hoped they would if the values werenât even close to what they need to beđ But weâre back babyyyyyy!!!! Next is the frame design and CAD!
Itâs been a while! In the mean time I have been working on the frame for the robot! I mainly did some prototyping for the motor mounts. Which went⌠letâs say âcould have gone betterâđ
I had to print multiple protorypes because the motors have a fairly irregular shape, and Iâm not good enough to do it first tryâŚđ
Also, because of some mistakes when saving, there sometimes missed parts on the print that I did actually change. Which caused some errors in my brain. Next up is the PCB and battery housing (the head of the robot)
I finally finished the CAD! Today I finally finished the 3D CAD model of my balancing robot called AXIS! I had some setbacks and difficulties along the way, but thatâs normal! Eventually I found a solution for all of them! Thatâs exactly what engineering is about! Letâs get printing Oh god⌠this will take daysđĽ
My parts arrived! Hi again! Itâs been a long long time since my last devlog. The reason⌠MY PARTSSS! They finally arrived, in the mean time I printed the entire frame and prepared everything. The great news is! I already soldered everything, it went so smoothly. Only things left are testing if my pcb works aandd integrating it into the frame! See you (hopefully in not that long of a time)
Ships:
I finished my balancin robot assembly! Today I finally finished my dual wheeled balancing robot that I spent such a long time making (mainly because of deliverytimes)! I did have to resolder my entire pcb because I soldered my motor driver on⌠upside down đŤŁBut all good now! I then integrated everything into the 3d printed frame. And voila, the robot is done!! ⌠for the hardware side Luckly I did make the code beforehand while waiting for deliveries and such! I also added a feature to steer the robot with my hands!!! (see video in the github âgalleryâ folder) I was very lucky that this code almost worked from the beginning, I only still need to tune my PID a bit more but since Stardance is ending I needed to ship nowđĽ I want to thank you all for following my journey here!!...
|
1793249364.9926393
|
1793335764.9926393
|
|
7475
|
My first slack bot! by @Mostafa | Stardance
|
Mostafa
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My first slack bot! by @Mostafa | Stardance
|
1792695282.0706484
|
1793300082.0706484
|
|
7476
|
Soundpad by @DDani | Stardance
|
DDani
|
0
|
2.1
|
0
|
[]
|
[{"post_id": 44623, "ts": "202 [{"post_id": 44623, "ts": "2026-08-10T14:54:20Z", "hours": 2.1, "text": "I have made my first custom hardware project, named Soundpad. It is a hackpad to control your media (skip, play, pause, prev), and has hotkeys of unused F-keys for other apps to bind."}]...
|
Title: Soundpad by @DDani | Stardance
Devlogs:
I Title: Soundpad by @DDani | Stardance
Devlogs:
I have made my first custom hardware project, named Soundpad. It is a hackpad to control your media (skip, play, pause, prev), and has hotkeys of unused F-keys for other apps to bind....
|
1793231421.2745225
|
1793836221.2745225
|
|
7477
|
My first slack bot! by @Zaptrack | Stardance
|
Zaptrack
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My first slack bot! by @Zaptrack | Stardanc Title: My first slack bot! by @Zaptrack | Stardance...
|
1792781401.7035775
|
1793386201.7035775
|
|
7478
|
Litter Detection Project by @AlwinT936 | Stardance
|
AlwinT936
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Litter Detection Project by @AlwinT936 | St Title: Litter Detection Project by @AlwinT936 | Stardance...
|
1792709668.9734483
|
1793314468.9734483
|
|
7479
|
My awesome macropad! by @soulcloude | Stardance
|
soulcloude
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @soulcloude | Stard Title: My awesome macropad! by @soulcloude | Stardance...
|
1793231447.0983582
|
1793836247.0983582
|
|
7480
|
My awesome macropad! by @batmobyle | Stardance
|
batmobyle
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @batmobyle | Starda Title: My awesome macropad! by @batmobyle | Stardance...
|
1793342907.7147813
|
1793947707.7147813
|
|
7481
|
Untitled by @soulcloude | Stardance
|
soulcloude
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Untitled by @soulcloude | Stardance
|
1792796045.5398774
|
1793400845.5398774
|
|
7482
|
My first slack bot! by @Spizza | Stardance
|
Spizza
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My first slack bot! by @Spizza | Stardance
|
1793249739.5824287
|
1793854539.5824287
|
|
7483
|
My first slack bot! by @soulcloude | Stardance
|
soulcloude
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My first slack bot! by @soulcloude | Starda Title: My first slack bot! by @soulcloude | Stardance...
|
1792752599.0967326
|
1793357399.0967326
|
|
7484
|
Exterstellar by @vejas | Stardance
|
vejas
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Exterstellar by @vejas | Stardance
|
1793252850.5494568
|
1793857650.5494568
|
|
7485
|
My awesome macropad! by @nadia_ | Stardance
|
nadia_
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @nadia_ | Stardance
|
1793260059.7409735
|
1793864859.7409735
|
|
7486
|
My awesome macropad! by @stewardtorerondahiro | St My awesome macropad! by @stewardtorerondahiro | Stardance...
|
stewardtorerondahiro
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @stewardtorerondahi Title: My awesome macropad! by @stewardtorerondahiro | Stardance...
|
1792738495.304169
|
1793343295.304169
|
|
7487
|
My awesome macropad! by @dpk_mocha13 | Stardance
|
dpk_mocha13
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @dpk_mocha13 | Star Title: My awesome macropad! by @dpk_mocha13 | Stardance...
|
1793350266.2427347
|
1793955066.2427347
|
|
7488
|
My first slack bot! by @BurntToAKrisp | Stardance
|
BurntToAKrisp
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My first slack bot! by @BurntToAKrisp | Sta Title: My first slack bot! by @BurntToAKrisp | Stardance...
|
1793220637.4170907
|
1793825437.4170907
|
|
7489
|
Weather app by @jupeme6 | Stardance
|
jupeme6
|
2
|
21.0
|
251
|
[{"ship_num": 1, "ts": "2026-0 [{"ship_num": 1, "ts": "2026-08-09T03:15:38Z", "hours": 7.0, "multiplier": 14.13, "payout": 93, "blessing": "neutral", "text": "What did you make? A weather app that lets you search the current weather of any city and get a 5-day forecast. Each search shows temperature, humidity, wind speed, feels-like, and a description, with a GIF background that matches the weather condition (sun, rain, snow, storm, etc.). Built in four betas: basic lookup, glassmorphism redesign with weather GIFs, 5-day forecast with a temperature chart drawn on canvas, and the final version with geolocation, favorites, PWA support, plus a settings panel with language (ES/EN) and units (Celsius/Fahrenheit) toggles added in response to reviewer feedback. The app is installable on desktop and mobile, works offline, and uses OpenWeatherMap + Giphy APIs. What was challenging? Two bugs almost broke me. My editorâs autocomplete kept rewriting value to ariaValueMax , which silently broke the search button in two projects. Later, when I tried to extend my mostrarClima function by saving the old one and redefining it, I caused infinite recursion because of JavaScript hoisting. Fix was to use a function expression instead. Also, the Giphy search API returned random GIFs that didnât represent the weather (got a dancing water drop for London), so I picked one fixed GIF per weather type by hand and hardcoded the URLs. The internationalization had a sneaky bug too. The language toggle translated the main UI but left the forecast section in Spanish because that code was hardcoded from an earlier beta. Had to re-query the API with the correct lang parameter and re-translate the forecast cards, the canvas day labels, and the legend separately. What are you proud of? The temperature chart is drawn with vanilla canvas, no Chart.js, just moveTo/lineTo/arc. About 60 lines of code, exactly the result I wanted. Also the weather GIF backgrounds: each weather type has a fixed GIF picked by hand, so it feels like youâre actually looking at the sky of that city. And the whole thing works as an installable PWA with offline support, plus now supports English and Fahrenheit for international users. What should people know so they can test your project? Open the demo URL and type a city name (Madrid, London, Tokyo, New York). Youâll see the current weather with a matching GIF background, plus a 5-day forecast with a temperature chart below. Click the gear icon in the top right to open settings and switch between Spanish / English and Celsius / Fahrenheit. Click the star next to the city name to add it to favorites. Click âMy locationâ to get weather for your actual coordinates. On Chrome you can install the app from the address bar icon. Live demo: https://notjuangamer10.github.io/weather-app/ Source: https://github.com/notjuangamer10/weather-app", "repo": "https://github.com/NOTJuangamer10/weather-app", "demo": "https://notjuangamer10.github.io/weather-app/"}, {"ship_num": 2, "ts": "2026-08-01T12:18:21Z", "hours": 14.0, "multiplier": 11.24, "payout": 158, "blessing": "neutral", "text": "What did you make? A weather app that lets you search the current weather of any city and get a 5-day forecast. Each search shows temperature, humidity, wind speed, feels-like, and a description, with a GIF background that matches the weather condition (sun, rain, snow, storm, etc.). Built in four betas: basic lookup, glassmorphism redesign with weather GIFs, 5-day forecast with a temperature chart drawn on canvas, and the final version with geolocation, favorites and PWA support. The app is installable on desktop and mobile, works offline, and uses OpenWeatherMap + Giphy APIs. What was challenging? Two bugs almost broke me. My editorâs autocomplete kept rewriting value to ariaValueMax , which silently broke the search button in two projects. Later, when I tried to extend my mostrarClima function by saving the old one and redefining it, I caused infinite recursion because of JavaScript hoisting - function declarations get hoisted, so I was saving the new function instead of the old one. Fix was to use a function expression instead. Also, the Giphy search API returned random GIFs that didnât represent the weather (got a dancing water drop for London), so I picked one fixed GIF per weather type by hand and hardcoded the URLs. What are you proud of? The temperature chart is drawn with vanilla canvas - no Chart.js, just moveTo/lineTo/arc. About 60 lines of code, exactly the result I wanted. Also the weather GIF backgrounds: instead of a static gradient or random GIFs, each weather type has a fixed GIF picked by hand, so it feels like youâre actually looking at the sky of that city. And the whole thing works as an installable PWA with offline support. What should people know so they can test your project? Open the demo URL and type a city name (Madrid, London, Tokyo, New York). Youâll see the current weather with a matching GIF background, plus a 5-day forecast with a temperature chart below. Click the star next to the city name to add it to favorites â they show up as chips and persist across reloads. Click âMy locationâ to get weather for your actual coordinates (browser will ask for permission). On Chrome you can install the app from the address bar icon, after that it opens in its own window and works offline. Live demo: https://notjuangamer10.github.io/weather-app/ Source: https://github.com/notjuangamer10/weather-app", "repo": "https://github.com/NOTJuangamer10/weather-app", "demo": "https://notjuangamer10.github.io/weather-app/"}]...
|
[{"post_id": 43697, "ts": "202 [{"post_id": 43697, "ts": "2026-08-09T03:14:25Z", "hours": 6.5617, "text": "Small patch for the Weather App based on reviewer feedback. Three fixes. First, the language toggle now actually translates the full UI. Before, switching to English left several strings in Spanish: the forecast heading, the day labels in both the cards and the canvas chart, the âhighs / lowsâ legend, and most embarrassingly, the weather description itself (âalgo de nubesâ instead of âscattered cloudsâ). The issue was that the forecast code was hardcoded in Spanish. Now it queries the OpenWeatherMap API with the correct lang parameter, and re-translates the forecast cards, the canvas labels, and the legend when the language changes. Second, the units toggle (Celsius / Fahrenheit) was already working but now also converts wind speed from km/h to mph when Fahrenheit is selected, which makes sense for imperial users. Third, the welcome state is in place. First-time visitors now see a pulsing sun icon and a short prompt instead of a blank page with just a search bar. Thanks to everyone who tested and left feedback. Three reviewers asked for multi-language support and one asked for unit conversion, so those were the top requested features."}, {"post_id": 38538, "ts": "2026-08-01T00:07:24Z", "hours": 3.4889, "text": "Beta 4, the final version of the Weather App, is up. Three new things: Geolocation: thereâs a âMy locationâ button next to the search bar that uses the browserâs geolocation API to get your coordinates, then queries OpenWeatherMap by lat/lon instead of city name. Useful when you donât know the exact city name or just want your local weather fast. Favorites: each result has a star icon next to the city name. Click it to add the city to your favorites, which show up as chips below the search bar. Click a chip to search that city again, click the Ă to remove it. Favorites are stored in localStorage so they survive reloads. PWA: added a manifest and a service worker, so the app is now installable on desktop and mobile. Once installed it opens in its own window without the browser chrome, and the service worker caches the main files so it works offline too. Also changed the font from system-ui to Space Grotesk. The default sans-serif felt a bit generic, Space Grotesk gives it more character, especially on the big temperature numbers. This is the last beta for this project. Calling Weather App done."}, {"post_id": 37533, "ts": "2026-07-30T10:47:19Z", "hours": 3.8489, "text": "Beta 3 of the Weather App is up. Now after the current weather you get a 5-day forecast with the day name, weather icon and max/min temperature for each day!!!(kinda usefull i know), plus a small chart below showing the max (red) and min (blue) temperatures across the week. The chart is drawn on a canvas with vanilla JS, no Chart.js or any library, just moveto/lineTo/arc. The tricky part was the forecast API. It returns a data point every 3 hours, so I had to filter and group them by day, taking the actual max and min of all the 3-hour readings inside each day. Otherwise the âmaxâ was just the temperature at a random hour, not the real daily high. I also had to redefine my existing mostrarClima function to also trigger the forecast fetch - I kept the old function in a variable and called it inside the new one, so the GIF background logic I had before still runs. Took me a while to figure that pattern out. Next version will probably be the last: geolocation + favorites + installable PWA. But i don`t really know if geolocation is a good idea Update:Hit a nasty bug in this version: every search returned âcity not foundâ even for Madrid, (which definitely works). Turns out the pattern I used to extend mostrarClima (saving the old function and redefining it) caused an infinite recursion because of JavaScript hoisting -function declarations get hoisted to the top, so when I saved âthe old functionâ I was actually saving the new one, which then called itself forever. Changed function mostrarClima() to mostrarClima = function() (function expression instead of declaration) and it fixed it. Took me way too long to figure out."}, {"post_id": 36722, "ts": "2026-07-28T23:22:38Z", "hours": 4.1825, "text": "Beta 2 of the Weather App is up. The UI got a real redesign: the background is now a GIF that matches the weather condition based on the info that OpenWeathermap gave me: a sunny sky for clear weather, falling rain for rainy cities, snow for cold ones, lightning for storms, etc. Cards use glassmorphism (frosted glass) so the GIF shows through while the text stays readable, with a dark overlay to keep contrast. Originally I called the Giphy API on every search and got a random GIF, but the results were inconsistent, for example, once I searched London and got a dancing water drop instead of actual rain. So I picked one fixed GIF per weather type by hand and hardcoded them. Now it feels like youâre actually looking at the sky of that city, same view every time. Next version will add a 5-day forecast below the current weather."}, {"post_id": 36048, "ts": "2026-07-27T21:28:29Z", "hours": 2.5794, "text": "First version of my Weather App is up. You type a city name, hit Search, and it pulls temperature, humidity, wind, feels-like and a short description from OpenWeatherMap. UI is intentionally plain cuz I wanted to make sure the API actually works before making it pretty. Fought my editorâs autocomplete again: it kept rewriting value to ariaValueMax , which silently broke the search button. Same bug as in my last project, should really disable that suggestion. Also took me a while to get the .then / .catch / .finally chain right, especially handling 404 when the city doesnât exist without crashing the page. Next version will be a proper redesign. Right now it works but itâs ugly."}]...
|
Title: Weather app by @jupeme6 | Stardance
Devlog Title: Weather app by @jupeme6 | Stardance
Devlogs:
First version of my Weather App is up. You type a city name, hit Search, and it pulls temperature, humidity, wind, feels-like and a short description from OpenWeatherMap. UI is intentionally plain cuz I wanted to make sure the API actually works before making it pretty. Fought my editorâs autocomplete again: it kept rewriting value to ariaValueMax , which silently broke the search button. Same bug as in my last project, should really disable that suggestion. Also took me a while to get the .then / .catch / .finally chain right, especially handling 404 when the city doesnât exist without crashing the page. Next version will be a proper redesign. Right now it works but itâs ugly.
Beta 2 of the Weather App is up. The UI got a real redesign: the background is now a GIF that matches the weather condition based on the info that OpenWeathermap gave me: a sunny sky for clear weather, falling rain for rainy cities, snow for cold ones, lightning for storms, etc. Cards use glassmorphism (frosted glass) so the GIF shows through while the text stays readable, with a dark overlay to keep contrast. Originally I called the Giphy API on every search and got a random GIF, but the results were inconsistent, for example, once I searched London and got a dancing water drop instead of actual rain. So I picked one fixed GIF per weather type by hand and hardcoded them. Now it feels like youâre actually looking at the sky of that city, same view every time. Next version will add a 5-day forecast below the current weather.
Beta 3 of the Weather App is up. Now after the current weather you get a 5-day forecast with the day name, weather icon and max/min temperature for each day!!!(kinda usefull i know), plus a small chart below showing the max (red) and min (blue) temperatures across the week. The chart is drawn on a canvas with vanilla JS, no Chart.js or any library, just moveto/lineTo/arc. The tricky part was the forecast API. It returns a data point every 3 hours, so I had to filter and group them by day, taking the actual max and min of all the 3-hour readings inside each day. Otherwise the âmaxâ was just the temperature at a random hour, not the real daily high. I also had to redefine my existing mostrarClima function to also trigger the forecast fetch - I kept the old function in a variable and called it inside the new one, so the GIF background logic I had before still runs. Took me a while to figure that pattern out. Next version will probably be the last: geolocation + favorites + installable PWA. But i don`t really know if geolocation is a good idea Update:Hit a nasty bug in this version: every search returned âcity not foundâ even for Madrid, (which definitely works). Turns out the pattern I used to extend mostrarClima (saving the old function and redefining it) caused an infinite recursion because of JavaScript hoisting -function declarations get hoisted to the top, so when I saved âthe old functionâ I was actually saving the new one, which then called itself forever. Changed function mostrarClima() to mostrarClima = function() (function expression instead of declaration) and it fixed it. Took me way too long to figure out.
Beta 4, the final version of the Weather App, is up. Three new things: Geolocation: thereâs a âMy locationâ button next to the search bar that uses the browserâs geolocation API to get your coordinates, then queries OpenWeatherMap by lat/lon instead of city name. Useful when you donât know the exact city name or just want your local weather fast. Favorites: each result has a star icon next to the city name. Click it to add the city to your favorites, which show up as chips below the search bar. Click a chip to search that city again, click the Ă to remove it. Favorites are stored in localStorage so they survive reloads. PWA: added a manifest and a service worker, so the app is now installable on desktop and mobile. Once installed it opens in its own window without the browser chrome, and the service worker caches the main files so it works offline too. Also changed the font from system-ui to Space Grotesk. The default sans-serif felt a bit generic, Space Grotesk gives it more character, especially on the big temperature numbers. This is the last beta for this project. Calling Weather App done.
Small patch for the Weather App based on reviewer feedback. Three fixes. First, the language toggle now actually translates the full UI. Before, switching to English left several strings in Spanish: the forecast heading, the day labels in both the cards and the canvas chart, the âhighs / lowsâ legend, and most embarrassingly, the weather description itself (âalgo de nubesâ instead of âscattered cloudsâ). The issue was that the forecast code was hardcoded in Spanish. Now it queries the OpenWeatherMap API with the correct lang parameter, and re-translates the forecast cards, the canvas labels, and the legend when the language changes. Second, the units toggle (Celsius / Fahrenheit) was already working but now also converts wind speed from km/h to mph when Fahrenheit is selected, which makes sense for imperial users. Third, the welcome state is in place. First-time visitors now see a pulsing sun icon and a short prompt instead of a blank page with just a search bar. Thanks to everyone who tested and left feedback. Three reviewers asked for multi-language support and one asked for unit conversion, so those were the top requested features.
Ships:
What did you make? A weather app that lets you search the current weather of any city and get a 5-day forecast. Each search shows temperature, humidity, wind speed, feels-like, and a description, with a GIF background that matches the weather condition (sun, rain, snow, storm, etc.). Built in four betas: basic lookup, glassmorphism redesign with weather GIFs, 5-day forecast with a temperature chart drawn on canvas, and the final version with geolocation, favorites and PWA support. The app is installable on desktop and mobile, works offline, and uses OpenWeatherMap + Giphy APIs. What was challenging? Two bugs almost broke me. My editorâs autocomplete kept rewriting value to ariaValueMax , which silently broke the search button in two projects. Later, when I tried to extend my mostrarClima function by saving the old one and redefining it, I caused infinite recursion because of JavaScript hoisting - function declarations get hoisted, so I was saving the new function instead of the old one. Fix was to use a function expression instead. Also, the Giphy search API returned random GIFs that didnât represent the weather (got a dancing water drop for London), so I picked one fixed GIF per weather type by hand and hardcoded the URLs. What are you proud of? The temperature chart is drawn with vanilla canvas - no Chart.js, just moveTo/lineTo/arc. About 60 lines of code, exactly the result I wanted. Also the weather GIF backgrounds: instead of a static gradient or random GIFs, each weather type has a fixed GIF picked by hand, so it feels like youâre actually looking at the sky of that city. And the whole thing works as an installable PWA with offline support. What should people know so they can test your project? Open the demo URL and type a city name (Madrid, London, Tokyo, New York). Youâll see the current weather with a matching GIF background, plus a 5-day forecast with a temperature chart below. Click the star next to the city name to add it to favorites â they show up as chips and persist across reloads. Click âMy locationâ to get weather for your actual coordinates (browser will ask for permission). On Chrome you can install the app from the address bar icon, after that it opens in its own window and works offline. Live demo: https://notjuangamer10.github.io/weather-app/ Source: https://github.com/notjuangamer10/weather-app
What did you make? A weather app that lets you search the current weather of any city and get a 5-day forecast. Each search shows temperature, humidity, wind speed, feels-like, and a description, with a GIF background that matches the weather condition (sun, rain, snow, storm, etc.). Built in four betas: basic lookup, glassmorphism redesign with weather GIFs, 5-day forecast with a temperature chart drawn on canvas, and the final version with geolocation, favorites, PWA support, plus a settings panel with language (ES/EN) and units (Celsius/Fahrenheit) toggles added in response to reviewer feedback. The app is installable on desktop and mobile, works offline, and uses OpenWeatherMap + Giphy APIs. What was challenging? Two bugs almost broke me. My editorâs autocomplete kept rewriting value to ariaValueMax , which silently broke the search button in two projects. Later, when I tried to extend my mostrarClima function by saving the old one and redefining it, I caused infinite recursion because of JavaScript hoisting. Fix was to use a function expression instead. Also, the Giphy search API returned random GIFs that didnât represent the weather (got a dancing water drop for London), so I picked one fixed GIF per weather type by hand and hardcoded the URLs. The internationalization had a sneaky bug too. The language toggle translated the main UI but left the forecast section in Spanish because that code was hardcoded from an earlier beta. Had to re-query the API with the correct lang parameter and re-translate the forecast cards, the canvas day labels, and the legend separately. What are you proud of? The temperature chart is drawn with vanilla canvas, no Chart.js, just moveTo/lineTo/arc. About 60 lines of code, exactly the result I wanted. Also the weather GIF backgrounds: each weather type has a fixed GIF picked by hand, so it feels like youâre actually looking at the sky of that city. And the whole thing works as an installable PWA with offline support, plus now supports English and Fahrenheit for international users. What should people know so they can test your project? Open the demo URL and type a city name (Madrid, London, Tokyo, New York). Youâll see the current weather with a matching GIF background, plus a 5-day forecast with a temperature chart below. Click the gear icon in the top right to open settings and switch between Spanish / English and Celsius / Fahrenheit. Click the star next to the city name to add it to favorites. Click âMy locationâ to get weather for your actual coordinates. On Chrome you can install the app from the address bar icon. Live demo: https://notjuangamer10.github.io/weather-app/ Source: https://github.com/notjuangamer10/weather-app...
|
1792878681.3015943
|
1793742681.3015943
|
|
7490
|
NASA A Picture Of The Day by @parthpthite | Starda NASA A Picture Of The Day by @parthpthite | Stardance...
|
parthpthite
|
0
|
3.86
|
0
|
[]
|
[{"post_id": 5293, "ts": "2026 [{"post_id": 5293, "ts": "2026-06-09T00:47:03Z", "hours": 0.8725, "text": "I did a whole bunch of changes to how it looks. I changed the fonts, the background, the size of the images, and made the buttons bigger and underneath the explanation. It wasnât hard per se, just took a lot of fidgeting with."}, {"post_id": 4643, "ts": "2026-06-08T01:24:27Z", "hours": 0.5117, "text": "So now I made it to where the image is on the left side of the screen and the text is next to the image on the right. I still need to fix sizing though. Iâll probably make it to where all images are the same size and the text also is the same or changes based on the size of the description to fit the frame. I wonder if thereâs a way to automatically adjust everything so that all components fit onto the users specific screen size. I also ran into this issue where because NASA uses UTC, even though itâs June 7th for me in ETC it was trying to load an image for June 8th which does not exist. So I had to add an extra thing to my error handling where if I receive a 404 error because I couldnât fetch the image, itâll go back to the previous days image."}, {"post_id": 4500, "ts": "2026-06-07T20:41:09Z", "hours": 0.3225, "text": "I got buttons to switch between the days. This was definitely the hardest thing Iâve done before because the Date stuff in JS was very confusing. I ended up creating two separate functions to go forward or backwards one day. Itâs still slow and clunky and doesnât look that good but I think this is a good point to stop and start working on visuals."}, {"post_id": 4460, "ts": "2026-06-07T19:43:24Z", "hours": 0.5092, "text": "Iâve gotten the title, date, picture, and description to all appear on the local host website. It wasnât hard, just basic HTML but its progress nonetheless. Next up I need to make a way for the user to move forward and backward through the days so they can see more than one image. Right now you have to zoom out to 75% to actually see everything so once I get most of my features implemented, Iâll make it look better. Iâm having a lot of fun!"}, {"post_id": 3905, "ts": "2026-06-07T00:35:31Z", "hours": 1.6467, "text": "Iâve started working on the first parts of the code. JavaScript and React is pretty new to me (I do know Java) so Iâm really just googling a lot and I watched â How to FETCH data from an API using JavaScript âŠď¸â by Bro Code. Iâm trying not to use AI and Iâve been successful so far but idk how long thatâll last."}]...
|
Title: NASA A Picture Of The Day by @parthpthite | Title: NASA A Picture Of The Day by @parthpthite | Stardance
Devlogs:
Iâve started working on the first parts of the code. JavaScript and React is pretty new to me (I do know Java) so Iâm really just googling a lot and I watched â How to FETCH data from an API using JavaScript âŠď¸â by Bro Code. Iâm trying not to use AI and Iâve been successful so far but idk how long thatâll last.
Iâve gotten the title, date, picture, and description to all appear on the local host website. It wasnât hard, just basic HTML but its progress nonetheless. Next up I need to make a way for the user to move forward and backward through the days so they can see more than one image. Right now you have to zoom out to 75% to actually see everything so once I get most of my features implemented, Iâll make it look better. Iâm having a lot of fun!
I got buttons to switch between the days. This was definitely the hardest thing Iâve done before because the Date stuff in JS was very confusing. I ended up creating two separate functions to go forward or backwards one day. Itâs still slow and clunky and doesnât look that good but I think this is a good point to stop and start working on visuals.
So now I made it to where the image is on the left side of the screen and the text is next to the image on the right. I still need to fix sizing though. Iâll probably make it to where all images are the same size and the text also is the same or changes based on the size of the description to fit the frame. I wonder if thereâs a way to automatically adjust everything so that all components fit onto the users specific screen size. I also ran into this issue where because NASA uses UTC, even though itâs June 7th for me in ETC it was trying to load an image for June 8th which does not exist. So I had to add an extra thing to my error handling where if I receive a 404 error because I couldnât fetch the image, itâll go back to the previous days image.
I did a whole bunch of changes to how it looks. I changed the fonts, the background, the size of the images, and made the buttons bigger and underneath the explanation. It wasnât hard per se, just took a lot of fidgeting with....
|
1793260402.762307
|
1793865202.762307
|
|
7491
|
My awesome macropad! by @Rachrok | Stardance
|
Rachrok
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @Rachrok | Stardanc Title: My awesome macropad! by @Rachrok | Stardance...
|
1793173882.8840299
|
1793778682.8840299
|
|
7492
|
Smart Weather Station by @arduinosio2021 | Stardan Smart Weather Station by @arduinosio2021 | Stardance...
|
arduinosio2021
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Smart Weather Station by @arduinosio2021 | Title: Smart Weather Station by @arduinosio2021 | Stardance...
|
1793170343.0211775
|
1793775143.0211775
|
|
7493
|
My awesome macropad! by @thenasaplusit | Stardance
|
thenasaplusit
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @thenasaplusit | St Title: My awesome macropad! by @thenasaplusit | Stardance...
|
1792727462.5689614
|
1793332262.5689614
|
|
7494
|
STM BLUE by @omruk5689 | Stardance
|
omruk5689
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: STM BLUE by @omruk5689 | Stardance
|
1792770774.2093296
|
1793375574.2093296
|
|
7495
|
My first slack bot! by @ryderpeterson1262 | Starda My first slack bot! by @ryderpeterson1262 | Stardance...
|
ryderpeterson1262
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My first slack bot! by @ryderpeterson1262 | Title: My first slack bot! by @ryderpeterson1262 | Stardance...
|
1793220790.8113456
|
1793825590.8113456
|
|
7496
|
soul browser by @soulcloude | Stardance
|
soulcloude
|
1
|
124.0
|
0
|
[{"ship_num": 1, "ts": "2026-0 [{"ship_num": 1, "ts": "2026-06-12T09:57:29Z", "hours": 124.0, "multiplier": null, "payout": 0, "blessing": "neutral", "text": "I built the Soul Browser! Itâs a high-performance, privacy-first web browser built natively for macOS. Instead of relying on bloated web-wrappers like Electron, I built the entire UI natively in SwiftUI and hooked it into the Chromium Embedded Framework (CEF) backend. This gives it the lightning-fast V8 JavaScript engine and Blink rendering, wrapped in a beautiful, macOS-native shell. The biggest challenge was wrestling with memory optimization and IPC (Inter-Process Communication). Because I avoided Electron, I had to manually manage how the native Swift UI communicated with the underlying Chromium processes. I spent a lot of time stripping down site isolation overhead and tuning the V8 heap limits so the browser stays incredibly lightweight. We also had to build custom retry-logic to stabilize our local AI Codex server connection! Iâm incredibly proud of the UI aesthetics and the performance metrics. The browser looks and feels like an expensive Apple product thanks to the frosted glassmorphism and native Swift animations. But more importantly, it runs like a dream. Features like the AI Tab Grouper and native !Bangs integration make it a genuine power-tool for developers. You can download the v1.0.0-beta directly from the Releases tab on GitHub! Since itâs a native macOS app in beta, you may need to right-click and hit âOpenâ to bypass the unverified developer warning. Try out the !w bangs in the address bar, or open 20 tabs to watch the AI Tab Grouper automatically organize them for you!", "repo": "https://github.com/soulcloude00/soul-browser", "demo": "https://github.com/soulcloude00/soul-browser/releases/tag/v1.0.1-beta"}]...
|
[{"post_id": 7376, "ts": "2026 [{"post_id": 7376, "ts": "2026-06-12T09:36:13Z", "hours": 9.7667, "text": "We just open-sourced the Soul Browser! đ I wanted to build a browser that feels incredibly fast, native to macOS, and respects your privacyâwithout relying on heavy Electron wrappers. đ The Stack: Built entirely in native Swift & SwiftUI for a buttery smooth UI, powered by the Chromium Embedded Framework (CEF) under the hood for world-class web rendering and V8 JavaScript performance. ⨠Core Features: ⢠!Bangs Integration: Instantly search your favorite sites directly from the address bar (e.g., type !w macOS to jump straight to Wikipedia). ⢠Built for Developers: First-class DevTools support natively integrated, plus aggressive memory optimization so your browser doesnât compete with your compiler for RAM. ⢠Privacy & Security First: Zero telemetry. No aggressive tracking. We baked in anti-phishing scanners, Tor-proxy support, and secure local keychain integration. If youâre tired of bloated web browsers, come check out the code, leave a â, or contribute! đ GitHub: https://github.com/soulcloude00/soul-browser Let me know what you think of the architecture! đ #swiftui #macos #opensource #buildinpublic"}, {"post_id": 6403, "ts": "2026-06-10T19:31:08Z", "hours": 14.0, "text": "done a full setting page for the browser took almost 14 hours , so basically its focused all all the use cases , developers , normal users and so many - planing to make a browser that no other browser has achieved"}, {"post_id": 2945, "ts": "2026-06-05T21:05:35Z", "hours": 0.5053, "text": "made an onboarding pages for the browser"}, {"post_id": 2687, "ts": "2026-06-05T11:17:35Z", "hours": 0.7861, "text": "added local ai assisted chat and navigation panel like comet browser on my browser it took lot of time lol"}, {"post_id": 2653, "ts": "2026-06-05T09:56:11Z", "hours": 13.3472, "text": "soul browser devlog 2 - creating a web page for the browser so that users can download from my website"}, {"post_id": 2372, "ts": "2026-06-04T20:35:41Z", "hours": 85.8667, "text": "crafting my own browser"}]...
|
Title: soul browser by @soulcloude | Stardance
De Title: soul browser by @soulcloude | Stardance
Devlogs:
crafting my own browser
soul browser devlog 2 - creating a web page for the browser so that users can download from my website
added local ai assisted chat and navigation panel like comet browser on my browser it took lot of time lol
made an onboarding pages for the browser
done a full setting page for the browser took almost 14 hours , so basically its focused all all the use cases , developers , normal users and so many - planing to make a browser that no other browser has achieved
We just open-sourced the Soul Browser! đ I wanted to build a browser that feels incredibly fast, native to macOS, and respects your privacyâwithout relying on heavy Electron wrappers. đ The Stack: Built entirely in native Swift & SwiftUI for a buttery smooth UI, powered by the Chromium Embedded Framework (CEF) under the hood for world-class web rendering and V8 JavaScript performance. ⨠Core Features: ⢠!Bangs Integration: Instantly search your favorite sites directly from the address bar (e.g., type !w macOS to jump straight to Wikipedia). ⢠Built for Developers: First-class DevTools support natively integrated, plus aggressive memory optimization so your browser doesnât compete with your compiler for RAM. ⢠Privacy & Security First: Zero telemetry. No aggressive tracking. We baked in anti-phishing scanners, Tor-proxy support, and secure local keychain integration. If youâre tired of bloated web browsers, come check out the code, leave a â, or contribute! đ GitHub: https://github.com/soulcloude00/soul-browser Let me know what you think of the architecture! đ #swiftui #macos #opensource #buildinpublic
Ships:
I built the Soul Browser! Itâs a high-performance, privacy-first web browser built natively for macOS. Instead of relying on bloated web-wrappers like Electron, I built the entire UI natively in SwiftUI and hooked it into the Chromium Embedded Framework (CEF) backend. This gives it the lightning-fast V8 JavaScript engine and Blink rendering, wrapped in a beautiful, macOS-native shell. The biggest challenge was wrestling with memory optimization and IPC (Inter-Process Communication). Because I avoided Electron, I had to manually manage how the native Swift UI communicated with the underlying Chromium processes. I spent a lot of time stripping down site isolation overhead and tuning the V8 heap limits so the browser stays incredibly lightweight. We also had to build custom retry-logic to stabilize our local AI Codex server connection! Iâm incredibly proud of the UI aesthetics and the performance metrics. The browser looks and feels like an expensive Apple product thanks to the frosted glassmorphism and native Swift animations. But more importantly, it runs like a dream. Features like the AI Tab Grouper and native !Bangs integration make it a genuine power-tool for developers. You can download the v1.0.0-beta directly from the Releases tab on GitHub! Since itâs a native macOS app in beta, you may need to right-click and hit âOpenâ to bypass the unverified developer warning. Try out the !w bangs in the address bar, or open 20 tabs to watch the AI Tab Grouper automatically organize them for you!...
|
1793278097.7649293
|
1793364497.7649293
|
|
7497
|
My first slack bot! by @Purdy | Stardance
|
Purdy
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My first slack bot! by @Purdy | Stardance
|
1793274725.4643621
|
1793879525.4643621
|
|
7498
|
My first slack bot! by @aryanonthenet | Stardance
|
aryanonthenet
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My first slack bot! by @aryanonthenet | Sta Title: My first slack bot! by @aryanonthenet | Stardance...
|
1792781582.9594278
|
1793386382.9594278
|
|
7499
|
Aquarium Auto-topup 2000 by @tanmay_avi | Stardanc Aquarium Auto-topup 2000 by @tanmay_avi | Stardance...
|
tanmay_avi
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Aquarium Auto-topup 2000 by @tanmay_avi | S Title: Aquarium Auto-topup 2000 by @tanmay_avi | Stardance...
|
1792760080.431785
|
1793364880.431785
|
|
7500
|
What if? by @NachoLibre | Stardance
|
NachoLibre
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: What if? by @NachoLibre | Stardance
|
1792723812.853463
|
1793328612.853463
|