|
16801
|
RoboBattle by @savva_belinsky | Stardance
|
savva_belinsky
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: RoboBattle by @savva_belinsky | Stardance
|
1793393615.9806595
|
1793998415.9806595
|
|
16802
|
My first slack bot! by @Suni | Stardance
|
Suni
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My first slack bot! by @Suni | Stardance
|
1793209867.208555
|
1793814667.208555
|
|
16803
|
Guitar Tuner by @Alexander_Wolf | Stardance
|
Alexander_Wolf
|
0
|
2.08
|
0
|
[]
|
[{"post_id": 6975, "ts": "2026 [{"post_id": 6975, "ts": "2026-06-11T18:22:18Z", "hours": 2.0792, "text": "Devlog #01 I tried to determine the guitar string played with the frequencies of a 2 second audio (wav file). There are many issues with the audio quality which causes the program to just think there’s no noise and fail. I tried many attempts to improve the quality as well as the volume, but this just worked for 3 out of the 6 open strings. Right now the goal is to reach the 6 open strings. After that I’ll go for the variations. If you have some suggestions or have done something like this before, I would gladly accept any help possible ;)"}]...
|
Title: Guitar Tuner by @Alexander_Wolf | Stardance Title: Guitar Tuner by @Alexander_Wolf | Stardance
Devlogs:
Devlog #01 I tried to determine the guitar string played with the frequencies of a 2 second audio (wav file). There are many issues with the audio quality which causes the program to just think there’s no noise and fail. I tried many attempts to improve the quality as well as the volume, but this just worked for 3 out of the 6 open strings. Right now the goal is to reach the 6 open strings. After that I’ll go for the variations. If you have some suggestions or have done something like this before, I would gladly accept any help possible ;)...
|
1793523245.176088
|
1794128045.176088
|
|
16804
|
My awesome macropad! by @Msol | Stardance
|
Msol
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @Msol | Stardance
|
1793814846.3745592
|
1794419646.3745592
|
|
16805
|
Thermocylcer by @thho1752 | Stardance
|
thho1752
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Thermocylcer by @thho1752 | Stardance
|
1793494534.8597636
|
1794099334.8597636
|
|
16806
|
My first slack bot! by @mbghg | Stardance
|
mbghg
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My first slack bot! by @mbghg | Stardance
|
1793436862.0146122
|
1794041662.0146122
|
|
16807
|
Frictionless by @Thimid | Stardance
|
Thimid
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Frictionless by @Thimid | Stardance
|
1793782164.2197828
|
1794386964.2197828
|
|
16808
|
guns.lol replica by @0c6a | Stardance
|
0c6a
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: guns.lol replica by @0c6a | Stardance
|
1793285811.7372813
|
1793890611.7372813
|
|
16809
|
Beam by @ct112 | Stardance
|
ct112
|
0
|
1.17
|
0
|
[]
|
[{"post_id": 7072, "ts": "2026 [{"post_id": 7072, "ts": "2026-06-11T20:57:22Z", "hours": 1.165, "text": "Devlog #1 Beam, a local network file transfer app which allows a phone to send files directly to a computer using a QR code and browser based upload page. Current Features: Built desktop application and GUI using PySide6 Used FastAPI and Uvicorn for server hosting Automatic LAN IP detection Automatic file saving and organization What’s next: File preview Clipboard Sync Mac/Linux compatibility"}]...
|
Title: Beam by @ct112 | Stardance
Devlogs:
Devlog Title: Beam by @ct112 | Stardance
Devlogs:
Devlog #1 Beam, a local network file transfer app which allows a phone to send files directly to a computer using a QR code and browser based upload page. Current Features: Built desktop application and GUI using PySide6 Used FastAPI and Uvicorn for server hosting Automatic LAN IP detection Automatic file saving and organization What’s next: File preview Clipboard Sync Mac/Linux compatibility...
|
1793249719.7310476
|
1793854519.7310476
|
|
16810
|
My very own WebOS by @GrantR | Stardance
|
GrantR
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My very own WebOS by @GrantR | Stardance
|
1793454839.254355
|
1794059639.254355
|
|
16811
|
My very own WebOS by @lukaseckar | Stardance
|
lukaseckar
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My very own WebOS by @lukaseckar | Stardanc Title: My very own WebOS by @lukaseckar | Stardance...
|
1793285761.6164548
|
1793890561.6164548
|
|
16812
|
college tinder by @maryamka | Stardance
|
maryamka
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: college tinder by @maryamka | Stardance
|
1793224131.294078
|
1793828931.294078
|
|
16813
|
Tracking App by @trevordotson92 | Stardance
|
trevordotson92
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Tracking App by @trevordotson92 | Stardance
|
1793281680.3807778
|
1793886480.3807778
|
|
16814
|
Racing Simulator by @AbdulRehmanShah | Stardance
|
AbdulRehmanShah
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Racing Simulator by @AbdulRehmanShah | Star Title: Racing Simulator by @AbdulRehmanShah | Stardance...
|
1793318185.7399957
|
1793922985.7399957
|
|
16815
|
Data Migrator App for Windows by @trevordotson92 | Data Migrator App for Windows by @trevordotson92 | Stardance...
|
trevordotson92
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Data Migrator App for Windows by @trevordot Title: Data Migrator App for Windows by @trevordotson92 | Stardance...
|
1793224536.0016074
|
1793829336.0016074
|
|
16816
|
The VectorWheel by @vivaan_khowal | Stardance
|
vivaan_khowal
|
0
|
11.15
|
0
|
[]
|
[{"post_id": 32439, "ts": "202 [{"post_id": 32439, "ts": "2026-07-21T17:00:38Z", "hours": 2.0547, "text": "THE CAD IS COMPLETE! For now. Took about 2 hrs to finish the rest. I added screw holes for the rivet nuts and any threaded inserts. The plan is to use M4, M5, and M6 screws for this build since I already have those screws lying around. I added bumpers to the front fenders for extra protection and drilled some extra holes in the aluminum extrusion to route the wires. As of now, the CAD should be complete, and I may edit it later once I get the parts."}, {"post_id": 32374, "ts": "2026-07-21T14:41:35Z", "hours": 1.6639, "text": "Been a while since I worked on this, but I made a ton of progress and did a ton of off-camera research. First of all, I decided to go from a 14s2p battery pack to a 16s2p pack because I didn’t realise that the hub motor is rated for 60V, but there wasn’t enough space in the compartment to store that big of a pack, so I had to spend some time in CAD to extend the overall length by 2 inches. A few other things I did: I decided to use aluminum sheets for the base; the footpads are now 3D printed instead of wooden (I might go back to wooden, not sure yet), and I decided to use both electronic footpads and a remote control for better and safer riding. Now there is just some polishing to do in CAD, and then its just building the actual thing."}, {"post_id": 20943, "ts": "2026-07-02T17:53:27Z", "hours": 1.4961, "text": "Made a lot of progress today. I found a 3D model for a scooter hub motor that had a very similar shaft to the actual OneWheel motor I was going to buy, so I used that model and did some detailing on the tires. I added LED strips to the front and back for nighttime riding and a latching button and barrel jack for charging and turning it on. Next up are the screw mountings for all the pieces, and I think that’s it for the CAD"}, {"post_id": 18161, "ts": "2026-06-29T05:47:03Z", "hours": 1.0606, "text": "Figured out the battery pack! Honestly went way smoother than I thought it would, and now we have a battery pack that fits, plus some extra room for the BMS and any wires to add. Took a bit of research to figure out a way to clear some extra space on the board without increasing weight or sacrificing strength. Next up is a few external circuit upgrades that I want to add that official OneWheels don’t have."}, {"post_id": 17993, "ts": "2026-06-28T23:19:00Z", "hours": 1.2881, "text": "Worked on most of the frame, finished a few parts like the electronics housing, the footpads, and the fenders. I also decided to use a 21700 custom battery pack for power. The plan is to use a 14s2p config, so 2 rows of 14 cells, which adds up to 28 cells. Not sure how to fit everything yet."}, {"post_id": 17729, "ts": "2026-06-28T17:16:49Z", "hours": 1.1636, "text": "I ended up pivoting back to a hub motor because I found this kit online with a hub motor, ESC, and a few other parts that are perfect for a OneWheel. I modeled mockups of the components based on the dimensions given on the website and came up with this basic design. Next up, I need to figure out the battery."}, {"post_id": 16375, "ts": "2026-06-26T17:24:16Z", "hours": 0.4778, "text": "So I found out that getting aluminum machined stators and rotors for a hub motor is super expensive, so instead I’m planning on using two 6.5” hoverboard hub motors joined together inside an official OneWheel tire for a custom hub motor. It doesn’t fit perfectly, but I think it’s still workable with some metal adapters or if I change the tire."}, {"post_id": 16023, "ts": "2026-06-26T00:31:09Z", "hours": 0.8525, "text": "Did a bit of detailing for the actual wheel. I designed a basis for the stator and the rotor for the motor. Measurements need work, but it’s a start. Also added a tread pattern on the tire."}, {"post_id": 15481, "ts": "2026-06-25T05:16:15Z", "hours": 1.0925, "text": "Finished a basis for the design. For the tire, I’m going to be using the official OneWheel Pint S Performance Tire. For the frame, the electronics casing will be made either with ABS or some other strong plastic or some kind of metal. I want to design my own hub motor from scratch (literally starting with a metal coil and magnets) since hub motors for this size are super expensive"}]...
|
Title: The VectorWheel by @vivaan_khowal | Stardan Title: The VectorWheel by @vivaan_khowal | Stardance
Devlogs:
Finished a basis for the design. For the tire, I’m going to be using the official OneWheel Pint S Performance Tire. For the frame, the electronics casing will be made either with ABS or some other strong plastic or some kind of metal. I want to design my own hub motor from scratch (literally starting with a metal coil and magnets) since hub motors for this size are super expensive
Did a bit of detailing for the actual wheel. I designed a basis for the stator and the rotor for the motor. Measurements need work, but it’s a start. Also added a tread pattern on the tire.
So I found out that getting aluminum machined stators and rotors for a hub motor is super expensive, so instead I’m planning on using two 6.5” hoverboard hub motors joined together inside an official OneWheel tire for a custom hub motor. It doesn’t fit perfectly, but I think it’s still workable with some metal adapters or if I change the tire.
I ended up pivoting back to a hub motor because I found this kit online with a hub motor, ESC, and a few other parts that are perfect for a OneWheel. I modeled mockups of the components based on the dimensions given on the website and came up with this basic design. Next up, I need to figure out the battery.
Worked on most of the frame, finished a few parts like the electronics housing, the footpads, and the fenders. I also decided to use a 21700 custom battery pack for power. The plan is to use a 14s2p config, so 2 rows of 14 cells, which adds up to 28 cells. Not sure how to fit everything yet.
Figured out the battery pack! Honestly went way smoother than I thought it would, and now we have a battery pack that fits, plus some extra room for the BMS and any wires to add. Took a bit of research to figure out a way to clear some extra space on the board without increasing weight or sacrificing strength. Next up is a few external circuit upgrades that I want to add that official OneWheels don’t have.
Made a lot of progress today. I found a 3D model for a scooter hub motor that had a very similar shaft to the actual OneWheel motor I was going to buy, so I used that model and did some detailing on the tires. I added LED strips to the front and back for nighttime riding and a latching button and barrel jack for charging and turning it on. Next up are the screw mountings for all the pieces, and I think that’s it for the CAD
Been a while since I worked on this, but I made a ton of progress and did a ton of off-camera research. First of all, I decided to go from a 14s2p battery pack to a 16s2p pack because I didn’t realise that the hub motor is rated for 60V, but there wasn’t enough space in the compartment to store that big of a pack, so I had to spend some time in CAD to extend the overall length by 2 inches. A few other things I did: I decided to use aluminum sheets for the base; the footpads are now 3D printed instead of wooden (I might go back to wooden, not sure yet), and I decided to use both electronic footpads and a remote control for better and safer riding. Now there is just some polishing to do in CAD, and then its just building the actual thing.
THE CAD IS COMPLETE! For now. Took about 2 hrs to finish the rest. I added screw holes for the rivet nuts and any threaded inserts. The plan is to use M4, M5, and M6 screws for this build since I already have those screws lying around. I added bumpers to the front fenders for extra protection and drilled some extra holes in the aluminum extrusion to route the wires. As of now, the CAD should be complete, and I may edit it later once I get the parts....
|
1793281974.1594744
|
1793886774.1594744
|
|
16817
|
My hackpad by @avycado13 | Stardance
|
avycado13
|
0
|
6.03
|
0
|
[]
|
[{"post_id": 84061, "ts": "202 [{"post_id": 84061, "ts": "2026-10-29T02:34:26Z", "hours": 0.2864, "text": "I made it smaller bcuz it was over 100mm before."}, {"post_id": 60589, "ts": "2026-09-03T01:22:57Z", "hours": 1.8608, "text": "fixed it up for shipping. thanks kian!"}, {"post_id": 22892, "ts": "2026-07-05T15:05:24Z", "hours": 1.9169, "text": "i started and finished the cad!"}, {"post_id": 22468, "ts": "2026-07-04T22:56:58Z", "hours": 0.7833, "text": "i finished my PCB!!!! yesterday and today i had a really crappy headache which turns out to be a fever"}, {"post_id": 20042, "ts": "2026-07-01T15:34:19Z", "hours": 1.1781, "text": "i started my schematic"}]...
|
Title: My hackpad by @avycado13 | Stardance
Devlo Title: My hackpad by @avycado13 | Stardance
Devlogs:
i started my schematic
i finished my PCB!!!! yesterday and today i had a really crappy headache which turns out to be a fever
i started and finished the cad!
fixed it up for shipping. thanks kian!
I made it smaller bcuz it was over 100mm before....
|
1793382762.7774704
|
1793987562.7774704
|
|
16818
|
My very own WebOS by @avycado13 | Stardance
|
avycado13
|
2
|
9.0
|
2
|
[{"ship_num": 1, "ts": "2026-0 [{"ship_num": 1, "ts": "2026-07-08T21:26:40Z", "hours": 1.0, "multiplier": 2.51, "payout": 2, "blessing": "neutral", "text": "I made an OS Kernel in the web! i also ported it to run natively and i want to have almost complete feautre parity between them!", "repo": "https://github.com/avycado13/webos", "demo": "http://finicky-knot.surge.sh/"}, {"ship_num": 2, "ts": "2026-07-01T01:01:21Z", "hours": 8.0, "multiplier": null, "payout": 0, "blessing": "neutral", "text": "I made an os that works in both wasm in the browser and in x86. it doesnt have a window manager yet but it has a console and a basic file system. I plan on adding more advanced functionality and maybe even a gui!", "repo": "https://github.com/avycado13/webos", "demo": "http://finicky-knot.surge.sh/"}]...
|
[{"post_id": 24979, "ts": "202 [{"post_id": 24979, "ts": "2026-07-08T21:25:26Z", "hours": 0.4697, "text": "random work to polish it up. not much changed"}, {"post_id": 19771, "ts": "2026-07-01T04:54:49Z", "hours": 0.3511, "text": "i added better readme"}, {"post_id": 19648, "ts": "2026-07-01T00:58:24Z", "hours": 2.2219, "text": "I added more browser support including a console in a canvas and a filesystem and shell. the filesystem is just a vector where each file is just a string for path and a vector of u8s for its bytes."}, {"post_id": 18696, "ts": "2026-06-29T21:39:11Z", "hours": 2.1514, "text": "i got it to work in the browser!! next i should add canvas rendering and also maybe some actual programs and a filesystem"}, {"post_id": 7286, "ts": "2026-06-12T04:33:38Z", "hours": 3.1797, "text": "I followed https://os.phil-opp.com/ and now i am going to try to get this to run on the web and add a basic shell"}]...
|
Title: My very own WebOS by @avycado13 | Stardance Title: My very own WebOS by @avycado13 | Stardance
Devlogs:
I followed https://os.phil-opp.com/ and now i am going to try to get this to run on the web and add a basic shell
i got it to work in the browser!! next i should add canvas rendering and also maybe some actual programs and a filesystem
I added more browser support including a console in a canvas and a filesystem and shell. the filesystem is just a vector where each file is just a string for path and a vector of u8s for its bytes.
i added better readme
random work to polish it up. not much changed
Ships:
I made an os that works in both wasm in the browser and in x86. it doesnt have a window manager yet but it has a console and a basic file system. I plan on adding more advanced functionality and maybe even a gui!
I made an OS Kernel in the web! i also ported it to run natively and i want to have almost complete feautre parity between them!...
|
1793703279.7427635
|
1793789679.7427635
|
|
16819
|
My awesome macropad! by @lashajananashvili | Stard My awesome macropad! by @lashajananashvili | Stardance...
|
lashajananashvili
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @lashajananashvili Title: My awesome macropad! by @lashajananashvili | Stardance...
|
1793224556.3421977
|
1793829356.3421977
|
|
16820
|
Orbital Mechanics Visualizer by @duaesakinahasan | Orbital Mechanics Visualizer by @duaesakinahasan | Stardance...
|
duaesakinahasan
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Orbital Mechanics Visualizer by @duaesakina Title: Orbital Mechanics Visualizer by @duaesakinahasan | Stardance...
|
1793465307.9804618
|
1794070107.9804618
|
|
16821
|
CaffeineOS_lite by @otzpt | Stardance
|
otzpt
|
1
|
6.0
|
0
|
[{"ship_num": 1, "ts": "2026-0 [{"ship_num": 1, "ts": "2026-06-24T20:31:21Z", "hours": 6.0, "multiplier": null, "payout": 0, "blessing": "neutral", "text": "I created CaffeineOS Lite, a WebOS that runs entirely in the browser. It features draggable windows, a functional terminal with commands like apt install, neofetch, and sudo. There’s also a file explorer with folder navigation and image preview, a login screen, a welcome window at startup, and a Start Menu with a Shut Down button. The design follows a caffeine theme and uses a dark Catppuccin color palette. Try it out; no password is needed. Just type any username and press Enter!", "repo": "https://github.com/otzpt/WebOS1", "demo": "https://webos1-otzpt.netlify.app/"}]...
|
[{"post_id": 14280, "ts": "202 [{"post_id": 14280, "ts": "2026-06-23T10:12:39Z", "hours": 2.2986, "text": "CaffeineOS Lite, Devlog #3 Welcome Screen & Start Menu With the core OS functioning, I focused on two features that enhance its completeness: a proper welcome screen and a useful Start Menu. The welcome screen opens automatically after login as a draggable window, similar to every other window in the OS. It features a coffee icon, the CaffeineOS Lite name, and a brief description. This small addition makes the first impression feel more intentional rather than just dropping the user straight onto the desktop. The Start Menu required more thought. It slides up from the taskbar when you click the Start button and closes when you click anywhere outside of it. Inside, it has an apps list on the left and a system panel on the right with a Shut Down button, which takes you back to the login screen, resetting the session. The entire design includes a frosted glass effect using backdrop-filter: blur, matching the taskbar’s style. Both features prompted me to consider state more carefully. I ensured that the Start Menu closes at the right moment and that the welcome window only opens once per session, not on every login attempt."}, {"post_id": 8914, "ts": "2026-06-14T20:49:44Z", "hours": 0.8917, "text": "Devlog 2 — File explorer with folder navigation and image viewer Added a functional “My Documents” app with a real file explorer. The root shows three folders (Wallpapers, Icons, Projects), and double-clicking navigates into them with a “..” button to go back. Double-clicking an image opens a draggable mini-window with the image preview, centered on screen , reusing the same drag/z-index logic from the main windows. Double-clicking the WebOS1 project entry opens its GitHub repo in a new tab. Under the hood, the file system is just a JS object (fileSystem) mapping folder names to arrays of items, which made it easy to add new folders/files without touching the rendering logic."}, {"post_id": 7711, "ts": "2026-06-12T21:28:35Z", "hours": 2.6731, "text": "Devlog 1 — Draggable windows, terminal, and login screen Built the main desktop environment for WebOS1. Windows can now be dragged around the desktop and bring themselves to the front when clicked (z-index). Added a custom login screen that asks for a username on startup, no passwords, just stored in memory for the session. The big feature today was a working terminal app with a real shell-like prompt (username@webos:~$). It supports help, whoami, ls, pwd, clear, sudo (denial message for fun), apt install with animated fake install output, and neofetch showing system info. Also replaced the default wallpaper with the classic Windows XP Bliss background and added custom SVG/icon assets for the desktop shortcuts."}]...
|
Title: CaffeineOS_lite by @otzpt | Stardance
Devl Title: CaffeineOS_lite by @otzpt | Stardance
Devlogs:
Devlog 1 — Draggable windows, terminal, and login screen Built the main desktop environment for WebOS1. Windows can now be dragged around the desktop and bring themselves to the front when clicked (z-index). Added a custom login screen that asks for a username on startup, no passwords, just stored in memory for the session. The big feature today was a working terminal app with a real shell-like prompt (username@webos:~$). It supports help, whoami, ls, pwd, clear, sudo (denial message for fun), apt install with animated fake install output, and neofetch showing system info. Also replaced the default wallpaper with the classic Windows XP Bliss background and added custom SVG/icon assets for the desktop shortcuts.
Devlog 2 — File explorer with folder navigation and image viewer Added a functional “My Documents” app with a real file explorer. The root shows three folders (Wallpapers, Icons, Projects), and double-clicking navigates into them with a “..” button to go back. Double-clicking an image opens a draggable mini-window with the image preview, centered on screen , reusing the same drag/z-index logic from the main windows. Double-clicking the WebOS1 project entry opens its GitHub repo in a new tab. Under the hood, the file system is just a JS object (fileSystem) mapping folder names to arrays of items, which made it easy to add new folders/files without touching the rendering logic.
CaffeineOS Lite, Devlog #3 Welcome Screen & Start Menu With the core OS functioning, I focused on two features that enhance its completeness: a proper welcome screen and a useful Start Menu. The welcome screen opens automatically after login as a draggable window, similar to every other window in the OS. It features a coffee icon, the CaffeineOS Lite name, and a brief description. This small addition makes the first impression feel more intentional rather than just dropping the user straight onto the desktop. The Start Menu required more thought. It slides up from the taskbar when you click the Start button and closes when you click anywhere outside of it. Inside, it has an apps list on the left and a system panel on the right with a Shut Down button, which takes you back to the login screen, resetting the session. The entire design includes a frosted glass effect using backdrop-filter: blur, matching the taskbar’s style. Both features prompted me to consider state more carefully. I ensured that the Start Menu closes at the right moment and that the welcome window only opens once per session, not on every login attempt.
Ships:
I created CaffeineOS Lite, a WebOS that runs entirely in the browser. It features draggable windows, a functional terminal with commands like apt install, neofetch, and sudo. There’s also a file explorer with folder navigation and image preview, a login screen, a welcome window at startup, and a Start Menu with a Shut Down button. The design follows a caffeine theme and uses a dark Catppuccin color palette. Try it out; no password is needed. Just type any username and press Enter!...
|
1793746526.956073
|
1793832926.956073
|
|
16822
|
Poopy by @kasarla_tejas | Stardance
|
kasarla_tejas
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Poopy by @kasarla_tejas | Stardance
|
1793778465.721123
|
1794383265.721123
|
|
16823
|
My very own WebOS by @Anyacado | Stardance
|
Anyacado
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My very own WebOS by @Anyacado | Stardance
|
1793537577.6491034
|
1794142377.6491034
|
|
16824
|
My awesome macropad! by @tsuki | Stardance
|
tsuki
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @tsuki | Stardance
|
1793418907.3548665
|
1794023707.3548665
|
|
16825
|
My first slack bot! by @romyox1337 | Stardance
|
romyox1337
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My first slack bot! by @romyox1337 | Starda Title: My first slack bot! by @romyox1337 | Stardance...
|
1793267742.1584032
|
1793872542.1584032
|
|
16826
|
Habit Tracker by @wingnovel | Stardance
|
wingnovel
|
1
|
22.0
|
38
|
[{"ship_num": 1, "ts": "2026-0 [{"ship_num": 1, "ts": "2026-07-06T17:35:22Z", "hours": 22.0, "multiplier": 2.19, "payout": 38, "blessing": "neutral", "text": "This project is a Habit tracker that helps you effectively understand your routine with smart insights to help you stay consistent in your habit formation. The challenging part in this project was figuring out how to make the UI look right. I struggle a lot with aesthetic choices. I am proud of the fact that all the code for this project was written by me even if to make the compiling more efficient i used AI to clump up my code into reusable functions for the repeated code scripts. All the requirements to run this program is install the exe file from the github repo.", "repo": "https://github.com/raghav03102008/habit_tracker", "demo": "https://github.com/raghav03102008/habit_tracker/releases/tag/v1"}]...
|
[{"post_id": 23594, "ts": "202 [{"post_id": 23594, "ts": "2026-07-06T17:23:22Z", "hours": 7.5383, "text": "In this session I fully completed the interface ready to ship. I have added all the pages. And ensured that all the pages store the data in the json file generated upon first launch of the app correctly."}, {"post_id": 7620, "ts": "2026-06-12T18:56:31Z", "hours": 14.7333, "text": "Building the sidebar and the dashboard with the right amount of padding and setting up the initial database. Deciding the colors and set up"}]...
|
Title: Habit Tracker by @wingnovel | Stardance
De Title: Habit Tracker by @wingnovel | Stardance
Devlogs:
Building the sidebar and the dashboard with the right amount of padding and setting up the initial database. Deciding the colors and set up
In this session I fully completed the interface ready to ship. I have added all the pages. And ensured that all the pages store the data in the json file generated upon first launch of the app correctly.
Ships:
This project is a Habit tracker that helps you effectively understand your routine with smart insights to help you stay consistent in your habit formation. The challenging part in this project was figuring out how to make the UI look right. I struggle a lot with aesthetic choices. I am proud of the fact that all the code for this project was written by me even if to make the compiling more efficient i used AI to clump up my code into reusable functions for the repeated code scripts. All the requirements to run this program is install the exe file from the github repo....
|
1793692216.7552905
|
1793778616.7552905
|
|
16827
|
My first slack bot! by @mokshith_brahmasani | Star My first slack bot! by @mokshith_brahmasani | Stardance...
|
mokshith_brahmasani
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My first slack bot! by @mokshith_brahmasani Title: My first slack bot! by @mokshith_brahmasani | Stardance...
|
1793411654.5087528
|
1794016454.5087528
|
|
16828
|
Vendora Lite Hub by @Titus-dev | Stardance
|
Titus-dev
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Vendora Lite Hub by @Titus-dev | Stardance
|
1793213268.3414958
|
1793818068.3414958
|
|
16829
|
My awesome macropad! by @0iez9n0sl | Stardance
|
0iez9n0sl
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @0iez9n0sl | Starda Title: My awesome macropad! by @0iez9n0sl | Stardance...
|
1793411526.521702
|
1794016326.521702
|
|
16830
|
My awesome macropad! by @armybajaj28 | Stardance
|
armybajaj28
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @armybajaj28 | Star Title: My awesome macropad! by @armybajaj28 | Stardance...
|
1793771672.0556438
|
1794376472.0556438
|
|
16831
|
My very own WebOS by @paranormalcaramel | Stardanc My very own WebOS by @paranormalcaramel | Stardance...
|
paranormalcaramel
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My very own WebOS by @paranormalcaramel | S Title: My very own WebOS by @paranormalcaramel | Stardance...
|
1793317826.6776214
|
1793922626.6776214
|
|
16832
|
Untitled project by @neszti | Stardance
|
neszti
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Untitled project by @neszti | Stardance
|
1793310958.893011
|
1793915758.893011
|
|
16833
|
My awesome macropad! by @berilcsr | Stardance
|
berilcsr
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @berilcsr | Stardan Title: My awesome macropad! by @berilcsr | Stardance...
|
1793361030.4216251
|
1793965830.4216251
|
|
16834
|
My awesome macropad! by @sargupta331 | Stardance
|
sargupta331
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @sargupta331 | Star Title: My awesome macropad! by @sargupta331 | Stardance...
|
1793465282.5666745
|
1794070082.5666745
|
|
16835
|
LoWa System by @kieran_parkinson | Stardance
|
kieran_parkinson
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: LoWa System by @kieran_parkinson | Stardanc Title: LoWa System by @kieran_parkinson | Stardance...
|
1793569724.1617286
|
1794174524.1617286
|
|
16836
|
My awesome macropad! by @Jaciel | Stardance
|
Jaciel
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @Jaciel | Stardance
|
1793400832.1943853
|
1794005632.1943853
|
|
16837
|
Grail-Tracker by @helicaptain | Stardance
|
helicaptain
|
1
|
2.35
|
0
|
[{"ship_num": 1, "ts": "2026-0 [{"ship_num": 1, "ts": "2026-06-12T04:21:14Z", "hours": 2.0, "multiplier": null, "payout": 0, "blessing": "neutral", "text": "🛠️ What did you make? I built Grail Tracker, which is a fast mobile app for tracking and cataloging high-value collectibles or milestones (your “grails”). Spreadsheets are super annoying to look at, so I made a clean card layout where you can see the item name, description, rarity tier, and timestamp all at a glance. It also saves everything directly to your device’s memory, so your list doesn’t disappear when you close the app. 🧠 What was challenging? The hardest part was definitely getting the local saving feature to work right without breaking the app’s layout. At first, when I tried passing data around, the UI completely glitched out—buttons were overlapping and clipping behind other elements on the screen. I had to figure out how to make async storage and JSON payloads load up properly on startup, while fixing a bunch of messy CSS flexbox rules so the screen scales normally. 🏆 What are you proud of? I’m proud of how I used a “vibe coding” setup with Cursor, Claude, and v0 to build this. I acted as the main builder and designer—I chose the style, put all the separate code pieces together, and handled the whole GitHub setup—but I used AI like an assistant to quickly generate the basic layouts and help me look up weird syntax errors. Doing it this way let me ship a fully working app with saving features way faster than doing it completely from scratch. 🧪 What should people know so they can test the project? Since the app saves everything locally on your device, the best way to see how it works is to watch the short video demo I attached! If you want to check out the actual files or see my hour-by-hour progress, you can look through the public GitHub repo.", "repo": "https://github.com/helicaptain827-alt/grail-tracker", "demo": "https://youtube.com/shorts/8JImJ8J0FfE"}]...
|
[{"post_id": 7574, "ts": "2026 [{"post_id": 7574, "ts": "2026-06-12T17:40:46Z", "hours": 0.3519, "text": "I made an app called Grail-Tracker. Where you can insert clothing items from your wardrobe and it will generate you outfits based on your preferences of color palettes, vibes, etc. Here is a Vercel link to my project: https://grail-tracker.vercel.app/ . There is no api or backend installed yet so there won’t be any computing, just a frontend."}, {"post_id": 6997, "ts": "2026-06-11T19:03:48Z", "hours": 1.8431, "text": "Grail-Tracker App (AI Clothing App)"}]...
|
Title: Grail-Tracker by @helicaptain | Stardance
Title: Grail-Tracker by @helicaptain | Stardance
Devlogs:
Grail-Tracker App (AI Clothing App)
I made an app called Grail-Tracker. Where you can insert clothing items from your wardrobe and it will generate you outfits based on your preferences of color palettes, vibes, etc. Here is a Vercel link to my project: https://grail-tracker.vercel.app/ . There is no api or backend installed yet so there won’t be any computing, just a frontend.
Ships:
🛠️ What did you make? I built Grail Tracker, which is a fast mobile app for tracking and cataloging high-value collectibles or milestones (your “grails”). Spreadsheets are super annoying to look at, so I made a clean card layout where you can see the item name, description, rarity tier, and timestamp all at a glance. It also saves everything directly to your device’s memory, so your list doesn’t disappear when you close the app. 🧠 What was challenging? The hardest part was definitely getting the local saving feature to work right without breaking the app’s layout. At first, when I tried passing data around, the UI completely glitched out—buttons were overlapping and clipping behind other elements on the screen. I had to figure out how to make async storage and JSON payloads load up properly on startup, while fixing a bunch of messy CSS flexbox rules so the screen scales normally. 🏆 What are you proud of? I’m proud of how I used a “vibe coding” setup with Cursor, Claude, and v0 to build this. I acted as the main builder and designer—I chose the style, put all the separate code pieces together, and handled the whole GitHub setup—but I used AI like an assistant to quickly generate the basic layouts and help me look up weird syntax errors. Doing it this way let me ship a fully working app with saving features way faster than doing it completely from scratch. 🧪 What should people know so they can test the project? Since the app saves everything locally on your device, the best way to see how it works is to watch the short video demo I attached! If you want to check out the actual files or see my hour-by-hour progress, you can look through the public GitHub repo....
|
1793786006.132401
|
1793872406.132401
|
|
16838
|
AlpineOS by @ashish-raj-singh | Stardance
|
ashish-raj-singh
|
2
|
23.0
|
72
|
[{"ship_num": 1, "ts": "2026-0 [{"ship_num": 1, "ts": "2026-09-24T08:32:33Z", "hours": 11.0, "multiplier": 6.36, "payout": 72, "blessing": "neutral", "text": "This is AlpineOS 2 (WebOS-2) upgraded from the previous version of AlpineOS 1 (WebOS-2) UPss Window animations are now smoother and more responsive. Introduced Widgets and are draggable anywhere on the desktop screen Truly FullScreen mode is here, previously the fullscreen was still not using the space behind the dock and menu bar at top, now it does Menu Bar improvements, now it has calender, wifi and a brightness slider which actually workd A new Settings app, which now has the personalization settings, and a new Widgets section where you can add widgets to your desktop Auto hide dock is now working, it simply hides the dock and shows on hovering th bottom of the desktop and it can be toggled in the settings app The Challenge? The main hurdle for me was understanding the actual OS, and its working as i hve tried to keep this project as realistic as i could understand other then that, the animations and drag and drop functionality was a bit challenging, so right now drag and drop us not supported as i have my exams coming up ans stardance is also going to end 30 sep to i dont have that much time The Excitement? this project was particularly exiting for me not because of the web technologies like used react heavily i have been a web dev for like a year main thing was the OS part, i have never worked on an OS before and this was a new experience for me, and i really loved it i tried to understand the real Operating System i leanred about scheduling algorithms like FCFS, SJF, memory management,how kernel works and communicated, how things are separated and modularized so much more What can you do? just right click on desktop and try to add a widget try to hide dock from the settings app, and hover at the bottom of the desktop and see the dock appear back try to minimize and maximize the windows and see the animations try increasing and decreasing the brightness from the menu bar", "repo": "https://github.com/ashish757/AlpineOS", "demo": "https://alpine-os-eight.vercel.app"}, {"ship_num": 2, "ts": "2026-06-13T12:16:11Z", "hours": 12.0, "multiplier": null, "payout": 0, "blessing": "neutral", "text": "I built AlpineOS, A fully functional, persistent, browser-based operating system, not just a CSS UI mockup it’s powered by a custom Redux “kernel” that actually manages background processes independently from the visual windows. live Activity Manager that tracks active PIDs and simulated CPU/Memory telemetry. It includes a persistent Virtual File System An draggable Window Manager Terminal engine that interacts with the File System Persistency , File and folders are persistennt and so it the state of the OS, you leave and come back just find that it has retained the state. The Challange? The architecture and state management were incredibly complex. Synchonizing the UI with the actual process mangement in the “kernal” Persistence, making the whole website persistence. Creatin an Task manager was fun and exciting, it felt like a real low level OS feature Excitement??? hell yeah my whole time I have been building, high-level websites and backend. I never tried a low level project, although this is not Low Level but it forced me to explore the architecture and concepts of hoow operating system works at Low level, how terminal connect with the file system. who doesn’t get excited by low level ?? Being able to selec multiple files and folder, not dragable yet, but soon will be What can i do? When you boot it up, don’t just click the icons and look and design and colors, stress test the ARCHITECTURE! Open the Terminal , use cd to change directory and try standard commands like mkdir, touch, ls, and interact with the file system Open the Activity Manager side-by-side with other apps to watch the active processes update and check the live charts (dummy) Right-click the desktop to open Personalization and change the wallpaper. Most importantly: Open a few windows, move them around the screen, and then hard refresh your browser or maybe just restart and Watch the persistence engine drop you right back where you left off.", "repo": "https://github.com/ashish757/AlpineOS", "demo": "https://alpine-os-eight.vercel.app"}]...
|
[{"post_id": 73046, "ts": "202 [{"post_id": 73046, "ts": "2026-09-24T08:31:29Z", "hours": 2.3342, "text": "Devlog 3 Launchpad as promised i have created the launchpad , but the serach functionlity is not working in it sorry i know its just s amll filter thing but i got my exmas this week, im in rush need to feactor the whole assets thing as i wasn initially importing the images but now i have moved it to the public and how just load them all updated the readme and added icon for the launchpad A lot can be done i think, but i am limited with this event ending 30 sep and also my exams are coming up SO BYE FOR NOWW, i really loved this project"}, {"post_id": 72817, "ts": "2026-09-23T19:47:31Z", "hours": 3.2872, "text": "Real OS true now this feels like a real operating system Widgets added widgets to the desktop, can be added from the settings right there are jsut widgets for clock and calander but same design can be used to simply add many more kinds of widgets Menu Bar menu bar now has the brightnes control slider and wifi options (obviously wifi are fake here) Dock can now be auto hidden from settings and only shows up when you hover in the bottom jsut like real macos Integrated the perosnalization in the settings app up next i will add a launch pad, some more little apps, fix a issue related to navigation in finder that i just found out"}, {"post_id": 71881, "ts": "2026-09-21T18:06:25Z", "hours": 1.4678, "text": "Devlog 2 i think i have laid a good foundation for process and window management i am now focusing on the UI and UX of the desktop environment and adding some more app, fetures, and more OS like fell Minimizing and maximizing windows implemented feature to minimize and maxiimize the windows maximizing feature existed but there were some bugs and it wasnt really fullscreen so i fixed that and now it works as expected improve Dock to open or focus the app by detecting wheter app is open or closed or minimized Started improving and adding animations used framer-motion to animate the appearance and disappearance of the windows later i will add transitions to other elements as well Created a settings app to handle the widgets , appearance and general setting all i one place also fixed an infinite render issue in ActivityManagerApp that was caused by the useEffect"}, {"post_id": 70213, "ts": "2026-09-18T18:31:33Z", "hours": 4.1772, "text": "Devlog 1 (after shipping webos1) so its been like 2 months since i last worked on this project so i just revised the whole codebase once Realism so when i decided to build this my main goal was to make it as realistic as possible by realistic i mean to make use of concepts that is used in real linux OS, not just UI but the actual inner workking What went down any application when opened in OS has two parts The window The process Like for notepad app, there is a wrapper window which has the close button and menu options and then there is the actual notepad process which handles things related to the creation and saving of the notes file what was the problem In real OS whenever we try to close a window, the window wrapper simply sends a closing signal to the actual process saying hey process user wants to close you for now. after this the process either saves logs, current changes or show user a confirmation dialog before closing but in my webOS i was jsut closign the process when user clicked the close button on the window wrapper, which is not realistic at all There are three kinds of closing signals that a process can receive SIGTERM SIGKILL SIGQUIT i will not get into the details of these signals but they mean different things , like SIGTERM is a request to terminate the process whereas SIGKILL forces the process to terminate. Solution so i basically added a new property to the window state to track if a window has requested to close the process process actively listen to such state change from the window and replies accordingly and closes the app performs tasks like askeing user to save changes, show warning dialog etc before closing the process there are still some bugs and a lot of things to improve in this thing but i will do that in the next devlog"}, {"post_id": 8020, "ts": "2026-06-13T11:46:24Z", "hours": 3.6642, "text": "#update number 0.0000033 The old man, Terminal Now terminal supports file system operations, can create and delete folders and files Textpad now it can open text files and edit those and save, can create new files as well Activity Manger , task manager Shows the currently running processes in the os, also gives and option to forcefully terminate any process, shows CPU and RAM usage (dummy but mindful of currently running processes) Menu bar menu bar now has specific and dynamic context menu options for each app although its not available yet for each app, but the logic is implemented, I just need to write json config for the ui to display Personalization Right click on desktop and click on Personalization to change wallpaper, will support more customization soon Boot added a cool mock up to mimic shutdown and booting process of OS,"}, {"post_id": 7578, "ts": "2026-06-12T17:50:11Z", "hours": 3.2947, "text": "DevLOG FOURRR What do we got this time well well I now have a complete basic file management system build, I can create text files, edit them, open them, save them in any specific location , delete it, terminal now can run more commands ‘’‘ls, mkdir, rm. touch, echo, clear, etc’’’’ context menus It was time taking and exhausting as every section of the operating system required different type of context menus. It was so much messed up. I ended up creating something central where a single config can decide and make a whole context menu, Desktop is ALIVE Long awaited something I wanted to do is to display what there is in the desktop directory directly onto the desktop after making the complete file system."}, {"post_id": 7422, "ts": "2026-06-12T11:20:38Z", "hours": 0.8694, "text": "Dev Log Threeeee Added a browser as of now only a few websites are accessible like wikipedia.org, other websites reject the connection in a iframe, I am looking for solution to this, I was thinking to maybe compile a rust model web browser using ecmascript and run a browser in browser, looks tough for limited time I have for this project , will definitely give it a shot ( if you guys got any idea let me know ) yayyyy File system is now persistent well despite coolest feature this was fairly easy redux-persistence saved the day, used indexDB as localstorage is synchronous and only allows 5MB of storage -> the files created in desktop folder now show up on the desktop on top of wallpaper NEXT GOAL add some more commands in terminal like ls, touch, mkdir, rm ability to create and save text files make files and folders drag-able (DREAMYYYY) personlization options"}, {"post_id": 7393, "ts": "2026-06-12T10:04:18Z", "hours": 0.6861, "text": "Dev Log 2 Implemented resizing and maximizing of windows (easypzy lemon squeeze) Added some dummy options in the menu bar (will soon make them functional too 😉) Improved UI of file manager (Finder)"}, {"post_id": 7245, "ts": "2026-06-12T02:47:35Z", "hours": 3.4803, "text": "AlpineOS, lightweight web os for study and space research(will be coming soon) The name inspiration come from the alpine images of docker “light weight” I am trying to build is os such that user can sort of study about space. as of now I have build some basic file system, dragable windows, I have setup basic architecture for applications, processes, and files and folders hope fully in my next DEVlog, I will have implemented /fixed resizable windows control panel file and folder persistence some cool apps on space theme preventing windows from going out of screen, will try to make terminal somewhat working improve the UI ability to minimize the windows drag and drop upload of files like text, pdf, word excel etc,"}]...
|
Title: AlpineOS by @ashish-raj-singh | Stardance
Title: AlpineOS by @ashish-raj-singh | Stardance
Devlogs:
AlpineOS, lightweight web os for study and space research(will be coming soon) The name inspiration come from the alpine images of docker “light weight” I am trying to build is os such that user can sort of study about space. as of now I have build some basic file system, dragable windows, I have setup basic architecture for applications, processes, and files and folders hope fully in my next DEVlog, I will have implemented /fixed resizable windows control panel file and folder persistence some cool apps on space theme preventing windows from going out of screen, will try to make terminal somewhat working improve the UI ability to minimize the windows drag and drop upload of files like text, pdf, word excel etc,
Dev Log 2 Implemented resizing and maximizing of windows (easypzy lemon squeeze) Added some dummy options in the menu bar (will soon make them functional too 😉) Improved UI of file manager (Finder)
Dev Log Threeeee Added a browser as of now only a few websites are accessible like wikipedia.org, other websites reject the connection in a iframe, I am looking for solution to this, I was thinking to maybe compile a rust model web browser using ecmascript and run a browser in browser, looks tough for limited time I have for this project , will definitely give it a shot ( if you guys got any idea let me know ) yayyyy File system is now persistent well despite coolest feature this was fairly easy redux-persistence saved the day, used indexDB as localstorage is synchronous and only allows 5MB of storage -> the files created in desktop folder now show up on the desktop on top of wallpaper NEXT GOAL add some more commands in terminal like ls, touch, mkdir, rm ability to create and save text files make files and folders drag-able (DREAMYYYY) personlization options
DevLOG FOURRR What do we got this time well well I now have a complete basic file management system build, I can create text files, edit them, open them, save them in any specific location , delete it, terminal now can run more commands ‘’‘ls, mkdir, rm. touch, echo, clear, etc’’’’ context menus It was time taking and exhausting as every section of the operating system required different type of context menus. It was so much messed up. I ended up creating something central where a single config can decide and make a whole context menu, Desktop is ALIVE Long awaited something I wanted to do is to display what there is in the desktop directory directly onto the desktop after making the complete file system.
#update number 0.0000033 The old man, Terminal Now terminal supports file system operations, can create and delete folders and files Textpad now it can open text files and edit those and save, can create new files as well Activity Manger , task manager Shows the currently running processes in the os, also gives and option to forcefully terminate any process, shows CPU and RAM usage (dummy but mindful of currently running processes) Menu bar menu bar now has specific and dynamic context menu options for each app although its not available yet for each app, but the logic is implemented, I just need to write json config for the ui to display Personalization Right click on desktop and click on Personalization to change wallpaper, will support more customization soon Boot added a cool mock up to mimic shutdown and booting process of OS,
Devlog 1 (after shipping webos1) so its been like 2 months since i last worked on this project so i just revised the whole codebase once Realism so when i decided to build this my main goal was to make it as realistic as possible by realistic i mean to make use of concepts that is used in real linux OS, not just UI but the actual inner workking What went down any application when opened in OS has two parts The window The process Like for notepad app, there is a wrapper window which has the close button and menu options and then there is the actual notepad process which handles things related to the creation and saving of the notes file what was the problem In real OS whenever we try to close a window, the window wrapper simply sends a closing signal to the actual process saying hey process user wants to close you for now. after this the process either saves logs, current changes or show user a confirmation dialog before closing but in my webOS i was jsut closign the process when user clicked the close button on the window wrapper, which is not realistic at all There are three kinds of closing signals that a process can receive SIGTERM SIGKILL SIGQUIT i will not get into the details of these signals but they mean different things , like SIGTERM is a request to terminate the process whereas SIGKILL forces the process to terminate. Solution so i basically added a new property to the window state to track if a window has requested to close the process process actively listen to such state change from the window and replies accordingly and closes the app performs tasks like askeing user to save changes, show warning dialog etc before closing the process there are still some bugs and a lot of things to improve in this thing but i will do that in the next devlog
Devlog 2 i think i have laid a good foundation for process and window management i am now focusing on the UI and UX of the desktop environment and adding some more app, fetures, and more OS like fell Minimizing and maximizing windows implemented feature to minimize and maxiimize the windows maximizing feature existed but there were some bugs and it wasnt really fullscreen so i fixed that and now it works as expected improve Dock to open or focus the app by detecting wheter app is open or closed or minimized Started improving and adding animations used framer-motion to animate the appearance and disappearance of the windows later i will add transitions to other elements as well Created a settings app to handle the widgets , appearance and general setting all i one place also fixed an infinite render issue in ActivityManagerApp that was caused by the useEffect
Real OS true now this feels like a real operating system Widgets added widgets to the desktop, can be added from the settings right there are jsut widgets for clock and calander but same design can be used to simply add many more kinds of widgets Menu Bar menu bar now has the brightnes control slider and wifi options (obviously wifi are fake here) Dock can now be auto hidden from settings and only shows up when you hover in the bottom jsut like real macos Integrated the perosnalization in the settings app up next i will add a launch pad, some more little apps, fix a issue related to navigation in finder that i just found out
Devlog 3 Launchpad as promised i have created the launchpad , but the serach functionlity is not working in it sorry i know its just s amll filter thing but i got my exmas this week, im in rush need to feactor the whole assets thing as i wasn initially importing the images but now i have moved it to the public and how just load them all updated the readme and added icon for the launchpad A lot can be done i think, but i am limited with this event ending 30 sep and also my exams are coming up SO BYE FOR NOWW, i really loved this project
Ships:
I built AlpineOS, A fully functional, persistent, browser-based operating system, not just a CSS UI mockup it’s powered by a custom Redux “kernel” that actually manages background processes independently from the visual windows. live Activity Manager that tracks active PIDs and simulated CPU/Memory telemetry. It includes a persistent Virtual File System An draggable Window Manager Terminal engine that interacts with the File System Persistency , File and folders are persistennt and so it the state of the OS, you leave and come back just find that it has retained the state. The Challange? The architecture and state management were incredibly complex. Synchonizing the UI with the actual process mangement in the “kernal” Persistence, making the whole website persistence. Creatin an Task manager was fun and exciting, it felt like a real low level OS feature Excitement??? hell yeah my whole time I have been building, high-level websites and backend. I never tried a low level project, although this is not Low Level but it forced me to explore the architecture and concepts of hoow operating system works at Low level, how terminal connect with the file system. who doesn’t get excited by low level ?? Being able to selec multiple files and folder, not dragable yet, but soon will be What can i do? When you boot it up, don’t just click the icons and look and design and colors, stress test the ARCHITECTURE! Open the Terminal , use cd to change directory and try standard commands like mkdir, touch, ls, and interact with the file system Open the Activity Manager side-by-side with other apps to watch the active processes update and check the live charts (dummy) Right-click the desktop to open Personalization and change the wallpaper. Most importantly: Open a few windows, move them around the screen, and then hard refresh your browser or maybe just restart and Watch the persistence engine drop you right back where you left off.
This is AlpineOS 2 (WebOS-2) upgraded from the previous version of AlpineOS 1 (WebOS-2) UPss Window animations are now smoother and more responsive. Introduced Widgets and are draggable anywhere on the desktop screen Truly FullScreen mode is here, previously the fullscreen was still not using the space behind the dock and menu bar at top, now it does Menu Bar improvements, now it has calender, wifi and a brightness slider which actually workd A new Settings app, which now has the personalization settings, and a new Widgets section where you can add widgets to your desktop Auto hide dock is now working, it simply hides the dock and shows on hovering th bottom of the desktop and it can be toggled in the settings app The Challenge? The main hurdle for me was understanding the actual OS, and its working as i hve tried to keep this project as realistic as i could understand other then that, the animations and drag and drop functionality was a bit challenging, so right now drag and drop us not supported as i have my exams coming up ans stardance is also going to end 30 sep to i dont have that much time The Excitement? this project was particularly exiting for me not because of the web technologies like used react heavily i have been a web dev for like a year main thing was the OS part, i have never worked on an OS before and this was a new experience for me, and i really loved it i tried to understand the real Operating System i leanred about scheduling algorithms like FCFS, SJF, memory management,how kernel works and communicated, how things are separated and modularized so much more What can you do? just right click on desktop and try to add a widget try to hide dock from the settings app, and hover at the bottom of the desktop and see the dock appear back try to minimize and maximize the windows and see the animations try increasing and decreasing the brightness from the menu bar...
|
1793706830.6002617
|
1793793230.6002617
|
|
16839
|
Frictionless by @aryan1 | Stardance
|
aryan1
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Frictionless by @aryan1 | Stardance
|
1793220565.7722118
|
1793825365.7722118
|
|
16840
|
HELA ENGINE by @Sithumjith | Stardance
|
Sithumjith
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: HELA ENGINE by @Sithumjith | Stardance
|
1793437072.7093832
|
1794041872.7093832
|
|
16841
|
AIplanner by @18roma2paris | Stardance
|
18roma2paris
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: AIplanner by @18roma2paris | Stardance
|
1793818209.4019384
|
1794423009.4019384
|
|
16842
|
My awesome macropad! by @eliq123 | Stardance
|
eliq123
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @eliq123 | Stardanc Title: My awesome macropad! by @eliq123 | Stardance...
|
1793224337.2265615
|
1793829137.2265615
|
|
16843
|
My awesome macropad! by @omar_talete | Stardance
|
omar_talete
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @omar_talete | Star Title: My awesome macropad! by @omar_talete | Stardance...
|
1793249502.5793016
|
1793854302.5793016
|
|
16844
|
My very own WebOS by @fiilipkooo | Stardance
|
fiilipkooo
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My very own WebOS by @fiilipkooo | Stardanc Title: My very own WebOS by @fiilipkooo | Stardance...
|
1793371798.4995484
|
1793976598.4995484
|
|
16845
|
My first slack bot! by @engerbuilder | Stardance
|
engerbuilder
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My first slack bot! by @engerbuilder | Star Title: My first slack bot! by @engerbuilder | Stardance...
|
1793206600.633915
|
1793811400.633915
|
|
16846
|
Makemore by @xerneas | Stardance
|
|
0
|
31.15
|
0
|
[]
|
[{"post_id": 32823, "ts": "202 [{"post_id": 32823, "ts": "2026-07-22T06:52:47Z", "hours": 10.4544, "text": "modern-lm: modernizing the architecture (RMSNorm, RoPE, SwiGLU) Starting from my GPT-2 124M reproduction and dragging the architecture from 2019 to 2026, one component at a time. The plan is a ladder: swap one thing, understand exactly what it does, keep going. This devlog covers the three architecture swaps. What changed: LayerNorm to RMSNorm, and dropped every bias. RMSNorm skips the mean-subtraction and only rescales by root-mean-square times a learned gain. The residual stream’s problem is magnitude drift, not offset, so re-centering was doing nothing. Half the ops, one fewer param vector per norm, and it sets up Muon later (which wants a model that is purely matrices plus a few gains). Learned position embeddings to RoPE. Deleted the wpe table entirely. Instead of adding a position vector, RoPE rotates the query and key feature pairs by a position-dependent angle, so attention scores end up depending on the relative distance between tokens, not their absolute index. GELU MLP to SwiGLU. Replaced the Linear to GELU to Linear block with a gated FFN: one branch computes features, a second branch (through SiLU) decides how much of each passes, then they multiply and project down. Went with 4x width, so it’s a bigger model, not a param-matched swap. Deliberate: I care more about a stronger model here than a clean ablation, and I can spend the compute. What was actually hard: The swaps are each ~20 to 40 lines. The trap was RoPE. It’s the classic silent bug: get the pairing convention wrong between the cos/sin cache and the apply function and the model still trains fine, it just quietly underperforms and you never find out. So I wrote a unit test instead of trusting the loss curve: shift the whole sequence by k positions and assert the attention scores between the same token pairs are unchanged. Rotation invariance either holds or it doesn’t (like an odometer). Next QK-norm, untied embeddings + zero-init prWSD schedule. Then the actual point: benchmark the modernized model against the untouched GPT-2 baseline (Row 0) at a fixed token budget on FineWeb-Edu, and see which Then scale the frozen recipe to ~500M."}, {"post_id": 29364, "ts": "2026-07-16T08:26:54Z", "hours": 4.6358, "text": "GPT-2 124M Reproduction Reproduced GPT-2 124M (the old 2019 model) from scratch, trained on ~10B tokens across 2x RTX 5090 in ~7 hours. Final val loss ~3.10. Dataset FineWeb-Edu sample-10BT, ~10B tokens GPT-2 BPE (tiktoken), vocab 50257 padded to 50304 100 uint16 shards (99 train, 1 val), ~19GB Streaming download + multiprocessing tokenizer — switching to imap_unordered with a bigger chunksize took tokenizing from ~2 hours to ~6 minutes Architecture GPT-2 124M: embd=768, block_size=1024, weight-tied embeddings (~124M params) 12 blocks, each with causal self-attention (12 heads, flash attention) + MLP (Linear → GELU → Linear, 4x) Pre-LayerNorm + residuals on both sublayers Training bf16 + torch.compile, AdamW, OneCycleLR cosine with warmup, grad clip 1.0 0.5M-token effective batch (524,288), held fixed via gradient accumulation DDP across 2x RTX 5090 on RunPod, 19,073 steps (~1 epoch), ~7 hours Checkpoint every 200 steps + auto-resume babysit script Results Val loss ~3.10 Coherent, topical English continuations. It’s a base model, so it continues a prompt rather than answering it . Shipped an inference script + Gradio demo, uploaded to Hugging Face What was actually hard The architecture was basically the char transformer from last devlog, just wider (124M vs ~10M) with BPE instead of 65 chars. The model was the easy part. The real work was scale engineering: sharding 10B tokens to stream off disk, gradient accumulation to hit a 0.5M-token batch on cards that OOM at batch 8, multi-GPU gradient sync with DDP, mixed precision, and checkpoint/resume so a 7-hour cloud run survives interruptions. That’s what turns a transformer into a trained LM."}, {"post_id": 15980, "ts": "2026-06-25T23:09:10Z", "hours": 11.1422, "text": "Tiny Shakespeare (decoder-only transformer) Dataset: Tiny Shakespeare, ~1.1M characters vocab 65 Architecture: GPT-style decoder token embedding + learned position embedding (embd=384) 6 transformer blocks, each: multi-head causal self-attention (8 heads, head_size=48) feed-forward (Linear → ReLU → Linear) pre-LayerNorm + residual connections on both sublayers final LayerNorm → linear to vocab block_size=64, dropout=0.2 Training: AdamW, lr=1e-4 ExponentialLR, gamma=0.99995 batch size 1024 5000 iters CUDA + torch.compile Results: train loss = 1.26 val loss = 1.52 Why it worked: Attention replaces flattening. Instead of mashing all 64 embeddings into one vector, each position attends to the relevant earlier positions directly. Sequence structure is preserved instead of thrown away — the exact failure mode of mlp3. The receptive field is the full block (64 chars) and every position can use all of it, not just a fixed local window. Residual connections + pre-LayerNorm let 6 blocks stack and actually train. Position embeddings give the model order information that a bag-of-embeddings MLP never had. Against the mlp3 goals: Goal was dev loss < 1.7 → hit 1.52. Goal was “generate actual words and sentences” → it now produces speaker tags, line breaks, and mostly real English words in Shakespearean cadence. It learned the SPEAKER:\\n dialogue form like mlp3 did, but this time it fills the dialogue with coherent structure instead of gibberish. The train/dev gap: 1.26 vs 1.52→ much tighter than mlp3’s 1.12/2.02. Dropout 0.2 + a context long enough to actually be useful means it’s learning real patterns, not memorizing short-range ones. Notes / loose ends: LR never decays much → gamma=0.99995 over 5000 steps only drops lr from 1e-4 to ~7.8e-5, so the schedule is nearly flat. Loss was still falling at step 5000; more iters or a higher/decaying lr would likely push val below 1.5. Next: Decay the LR properly (or warmup + cosine) and train longer to close in on ~1.4. Scale block_size for longer context now that attention makes it affordable. Develop gpt-2 (the old sucky model from 2019)"}, {"post_id": 7016, "ts": "2026-06-11T19:28:20Z", "hours": 4.9217, "text": "makemore — devlog mlp2-gpu: names MLP Dataset: Karpathy’s names.txt (32k names, vocab 27) Architecture: embed=10 -> flatten -> linear(200) -> tanh -> linear(27) Training: AdamW, lr=1e-3 200 epochs MPS Result: dev loss = 2.20 Samples: bhuza fremah mykeslanna talee The model worked because names are short and mostly depend on local character patterns. mlp3: Tiny Shakespeare (without a guide and with single hidden layer) Dataset: 1.1M characters vocab 65 Architecture: embed=32, hidden=400, block_size=16 Training: batch size 8192 600 epochs CUDA Results: train loss = 1.12 dev loss = 2.02 Samples: LERCUTIF: MARCAPUL: HENRY: Why it failed: 16 characters of context is too short for prose. Flattening embeddings throws away sequence structure. The model learned the easiest pattern: SPEAKER:\\n dialogue. Large train/dev gap suggests overfitting to short-range patterns. The model was not too small. It was the wrong architecture. Next: WaveNet-style hierarchical MLP from the next lecture. Instead of flattening all embeddings at once, combine them in stages to preserve locality and expand the receptive field. Goal: dev loss < 1.7 on Tiny Shakespeare generate actual words and sentences After that: transformers."}]...
|
Title: Makemore by @xerneas | Stardance
Devlogs:
Title: Makemore by @xerneas | Stardance
Devlogs:
makemore — devlog mlp2-gpu: names MLP Dataset: Karpathy’s names.txt (32k names, vocab 27) Architecture: embed=10 -> flatten -> linear(200) -> tanh -> linear(27) Training: AdamW, lr=1e-3 200 epochs MPS Result: dev loss = 2.20 Samples: bhuza fremah mykeslanna talee The model worked because names are short and mostly depend on local character patterns. mlp3: Tiny Shakespeare (without a guide and with single hidden layer) Dataset: 1.1M characters vocab 65 Architecture: embed=32, hidden=400, block_size=16 Training: batch size 8192 600 epochs CUDA Results: train loss = 1.12 dev loss = 2.02 Samples: LERCUTIF: MARCAPUL: HENRY: Why it failed: 16 characters of context is too short for prose. Flattening embeddings throws away sequence structure. The model learned the easiest pattern: SPEAKER:\n dialogue. Large train/dev gap suggests overfitting to short-range patterns. The model was not too small. It was the wrong architecture. Next: WaveNet-style hierarchical MLP from the next lecture. Instead of flattening all embeddings at once, combine them in stages to preserve locality and expand the receptive field. Goal: dev loss < 1.7 on Tiny Shakespeare generate actual words and sentences After that: transformers.
Tiny Shakespeare (decoder-only transformer) Dataset: Tiny Shakespeare, ~1.1M characters vocab 65 Architecture: GPT-style decoder token embedding + learned position embedding (embd=384) 6 transformer blocks, each: multi-head causal self-attention (8 heads, head_size=48) feed-forward (Linear → ReLU → Linear) pre-LayerNorm + residual connections on both sublayers final LayerNorm → linear to vocab block_size=64, dropout=0.2 Training: AdamW, lr=1e-4 ExponentialLR, gamma=0.99995 batch size 1024 5000 iters CUDA + torch.compile Results: train loss = 1.26 val loss = 1.52 Why it worked: Attention replaces flattening. Instead of mashing all 64 embeddings into one vector, each position attends to the relevant earlier positions directly. Sequence structure is preserved instead of thrown away — the exact failure mode of mlp3. The receptive field is the full block (64 chars) and every position can use all of it, not just a fixed local window. Residual connections + pre-LayerNorm let 6 blocks stack and actually train. Position embeddings give the model order information that a bag-of-embeddings MLP never had. Against the mlp3 goals: Goal was dev loss < 1.7 → hit 1.52. Goal was “generate actual words and sentences” → it now produces speaker tags, line breaks, and mostly real English words in Shakespearean cadence. It learned the SPEAKER:\n dialogue form like mlp3 did, but this time it fills the dialogue with coherent structure instead of gibberish. The train/dev gap: 1.26 vs 1.52→ much tighter than mlp3’s 1.12/2.02. Dropout 0.2 + a context long enough to actually be useful means it’s learning real patterns, not memorizing short-range ones. Notes / loose ends: LR never decays much → gamma=0.99995 over 5000 steps only drops lr from 1e-4 to ~7.8e-5, so the schedule is nearly flat. Loss was still falling at step 5000; more iters or a higher/decaying lr would likely push val below 1.5. Next: Decay the LR properly (or warmup + cosine) and train longer to close in on ~1.4. Scale block_size for longer context now that attention makes it affordable. Develop gpt-2 (the old sucky model from 2019)
GPT-2 124M Reproduction Reproduced GPT-2 124M (the old 2019 model) from scratch, trained on ~10B tokens across 2x RTX 5090 in ~7 hours. Final val loss ~3.10. Dataset FineWeb-Edu sample-10BT, ~10B tokens GPT-2 BPE (tiktoken), vocab 50257 padded to 50304 100 uint16 shards (99 train, 1 val), ~19GB Streaming download + multiprocessing tokenizer — switching to imap_unordered with a bigger chunksize took tokenizing from ~2 hours to ~6 minutes Architecture GPT-2 124M: embd=768, block_size=1024, weight-tied embeddings (~124M params) 12 blocks, each with causal self-attention (12 heads, flash attention) + MLP (Linear → GELU → Linear, 4x) Pre-LayerNorm + residuals on both sublayers Training bf16 + torch.compile, AdamW, OneCycleLR cosine with warmup, grad clip 1.0 0.5M-token effective batch (524,288), held fixed via gradient accumulation DDP across 2x RTX 5090 on RunPod, 19,073 steps (~1 epoch), ~7 hours Checkpoint every 200 steps + auto-resume babysit script Results Val loss ~3.10 Coherent, topical English continuations. It’s a base model, so it continues a prompt rather than answering it . Shipped an inference script + Gradio demo, uploaded to Hugging Face What was actually hard The architecture was basically the char transformer from last devlog, just wider (124M vs ~10M) with BPE instead of 65 chars. The model was the easy part. The real work was scale engineering: sharding 10B tokens to stream off disk, gradient accumulation to hit a 0.5M-token batch on cards that OOM at batch 8, multi-GPU gradient sync with DDP, mixed precision, and checkpoint/resume so a 7-hour cloud run survives interruptions. That’s what turns a transformer into a trained LM.
modern-lm: modernizing the architecture (RMSNorm, RoPE, SwiGLU) Starting from my GPT-2 124M reproduction and dragging the architecture from 2019 to 2026, one component at a time. The plan is a ladder: swap one thing, understand exactly what it does, keep going. This devlog covers the three architecture swaps. What changed: LayerNorm to RMSNorm, and dropped every bias. RMSNorm skips the mean-subtraction and only rescales by root-mean-square times a learned gain. The residual stream’s problem is magnitude drift, not offset, so re-centering was doing nothing. Half the ops, one fewer param vector per norm, and it sets up Muon later (which wants a model that is purely matrices plus a few gains). Learned position embeddings to RoPE. Deleted the wpe table entirely. Instead of adding a position vector, RoPE rotates the query and key feature pairs by a position-dependent angle, so attention scores end up depending on the relative distance between tokens, not their absolute index. GELU MLP to SwiGLU. Replaced the Linear to GELU to Linear block with a gated FFN: one branch computes features, a second branch (through SiLU) decides how much of each passes, then they multiply and project down. Went with 4x width, so it’s a bigger model, not a param-matched swap. Deliberate: I care more about a stronger model here than a clean ablation, and I can spend the compute. What was actually hard: The swaps are each ~20 to 40 lines. The trap was RoPE. It’s the classic silent bug: get the pairing convention wrong between the cos/sin cache and the apply function and the model still trains fine, it just quietly underperforms and you never find out. So I wrote a unit test instead of trusting the loss curve: shift the whole sequence by k positions and assert the attention scores between the same token pairs are unchanged. Rotation invariance either holds or it doesn’t (like an odometer). Next QK-norm, untied embeddings + zero-init prWSD schedule. Then the actual point: benchmark the modernized model against the untouched GPT-2 baseline (Row 0) at a fixed token budget on FineWeb-Edu, and see which Then scale the frozen recipe to ~500M....
|
1792914540.4782085
|
1793778540.4782085
|
|
16847
|
My awesome macropad! by @kero | Stardance
|
kero
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My awesome macropad! by @kero | Stardance
|
1793321312.3473213
|
1793926112.3473213
|
|
16848
|
Wind Tunnel by @dragonbiteSniper | Stardance
|
dragonbiteSniper
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Wind Tunnel by @dragonbiteSniper | Stardanc Title: Wind Tunnel by @dragonbiteSniper | Stardance...
|
1793231606.6243553
|
1793836406.6243553
|
|
16849
|
My first slack bot! by @muhammadN | Stardance
|
muhammadN
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: My first slack bot! by @muhammadN | Stardan Title: My first slack bot! by @muhammadN | Stardance...
|
1793202616.7635717
|
1793807416.7635717
|
|
16850
|
Holy Tequila by @Un-Average-Human | Stardance
|
Un-Average-Human
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: Holy Tequila by @Un-Average-Human | Stardan Title: Holy Tequila by @Un-Average-Human | Stardance...
|
1793476073.8615303
|
1794080873.8615303
|