|
16981
|
iPod Touch 4 Jailbreak by @itech | Stardance
|
itech
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: iPod Touch 4 Jailbreak by @itech | Stardanc Title: iPod Touch 4 Jailbreak by @itech | Stardance...
|
1793368040.7272131
|
1793972840.7272131
|
|
58078
|
iTab by @PiGuy141 | Stardance
|
PiGuy141
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: iTab by @PiGuy141 | Stardance
|
1793469403.602741
|
1794074203.602741
|
|
36469
|
iTab: a 2013 iPad with Android and OLED by @water iTab: a 2013 iPad with Android and OLED by @water | Stardance...
|
water
|
0
|
19.0
|
0
|
[]
|
[{"post_id": 35284, "ts": "202 [{"post_id": 35284, "ts": "2026-07-26T16:20:56Z", "hours": 4.0, "text": "iTab DEVLOG-5!! not a revolutionary devlog like the last one byt still, many things have changed, and this time, i’ve finally got to test it out and share the results!!🥹 if you’re not intrested in the latest changes i’ve done, you can skip to the end and check out the test results! Improved the battery pack! if you seen one of my earlier devolgs, you’d probably know that i made a battery pack for the tablet with 2 phone batteries, that idea is fine, it’s cheap and offers an even higher capacity than the original battery! but who couldve guessed that wrapping an already thick part in 5 rounds in electrical tape would make it any thicker , i just unwrapped it and made sure its insulated. Reverted some changes!! i didnt really need the headphone jack nor the data pins so i removed them, i also got rid of the ZIF connector and just soldered the wires directly!, i know its kinda lazy and defeats alot of features but i genuinely dont need these features since all i’m using the tablet for is watching and studying, and they made the assembly so much harder so they had to go unfourtunately Redone all the button logic! Samsung’s a bitch they made the power button completely insulated from ground, meaning instead of needing to only run 3 wires for button signals ans grouping the ground, i had to run an extra wire JUST for the power button , while i was at it, i also mapped out what each pin connected to Tested out!!!! since the start of the project, i haven’t been able to test it out, but now, i am!!, here are what works and what doesn’t!: What works: Power button Volume up button Bluetooth Mic Basic functions Battery Charging (ignore the temp error the battery connector is just loose) What doesn’t work: Wi-Fi (Signal very low, since the tablet used to rely on it’s metal frame as an antenna) Volume down button (Surprisingly) USB-C Data (Didn’t wire them on purpose) Speakers (Very surprisingly, i thought i wired them perfectly The conclusion i’d say we’re ~65% into the project, i still need to fix all the problems i mentioned, and make everything easily accessible before we move to the last (and frankly the hardest) step, which is delaminating and replacing the OLED’s glass, and until then, stay goated gng✌"}, {"post_id": 34234, "ts": "2026-07-24T19:22:42Z", "hours": 10.0, "text": "iTab DEVLOG-4!! i haven’t posted anything in the past 3 days bc i was cooking up smth big , ive done so much to this project so far and i can say we’re halfway through it, here is everything i’ve done in the past 3 days: Made screw holes for the bracket to screw to the iPad frame: Since the tablet’s bracket was just sitting lose on the frame, i made some screw holes in frame out of (i kid you not) sim cards plastic placeholders , sounds weird ik but it works and sticks to metal very well since its just thick plastic, plus it elevated the bracket just enough to make it sit flush with the frame Added USB-C!!! Now THIS is a big one, i sourced a USB-C breakout board from a local repair shop, i was planning on going through the lengthy process of using a full Type-C port and adding resistors to the CC pins and shi, but i got lucky , i used superglue and hot glue to secure it in place and it suprisingly looks very good!! Wired all the I/O Power button, vol buttons, headphone jack, and Type-C connector, i soldered wires directly to them and routed them with electical tape, i also commoned all ground to the metal chasis Made a removable connector for the I/O!! in addition to wiring and routing I/O, i couldn’t just solder everything together since i know DAMN WELL i something will break and i’d find myself reslodering wires same (if not less) thickness than my hair, so i used a ZIF connector (or whatever it’s called) i pulled from a dead laptop to connect both sides, and i had to solder BOTH sides with of the connector with horrible lighting AND with a $3 iron i had since 2023 AND WITHOUT a microscope , let that sink in.. other than that i made a few minor improvments like adding a placebo camera lens for looks (since i’m NOT rewiring the camera from center to side without a microscope) and also made sure everything is routed neatly, i used specifically wires from old headphones since they were thin and insulated. Anyways that’s all for today!, i’d say we’re 50-55% through the project, byt he end of this week we will hopefully have finished it, i’ve also noticed this project going viral! thanks for all the support, and until next time, stay goated gng✌"}, {"post_id": 31532, "ts": "2026-07-20T00:39:17Z", "hours": 2.0, "text": "iTab DEVLOG-3!!!: A few important things were done today!! i’m very happy with how quickly this is coming together🥹 Made the battery pack!: yes you read that right, “made”, you see, the original battery was bloated and since it was old and expensive, i made my own , i pulled 2 almost identical li-ion cells from phones (4000mAh each) and wired them to the original bms which wires them in parallel, i then just wrapped them around electrical tape and now have a cool little 8000mAh 3.85v cell! Removed more off the bracket: Now the boring stuff, like the last 2 devlogs, i dremeled the bracket and the housing properly this time since i got an actual good dremel (borrowed but still , sanded the part that needs sanding more, and yeah now it fits perfectly!, the only thing remaining is making screw holes and i should move on to the motherboard and charging port and buttons…etc Aaaand yea thats all for today, tomorrow will be even more lit than today, until then stay goated gng✌"}, {"post_id": 31116, "ts": "2026-07-19T07:56:34Z", "hours": 1.0, "text": "iTab DEVLOG-2!!!: i haven’t done much today, but still some progress nonetheless! Cut more off the display bracket: at the bottom where the speakers are and on top where the buttons are, i cut more parts of the bracket, the cuts were rough since my only dremel started smoking the moment i turned it on so i had to only use pliers>:( Dremeled off the iPad housing: I removed unnecessary aluminium especially off the lte antenna (which was blocking most of the upper bracket) not much has been done today since i was tired but i’ll lock in and hopefully finish ts by the end of the week!! until then stay goated gng✌"}, {"post_id": 30282, "ts": "2026-07-17T18:38:36Z", "hours": 2.0, "text": "iTab DEVLOG-1!!! Today we lit🔥🔥🔥 i’ve done multiple improvements to this project!: Disassembled both tablets: ik, boring but figured i’d mention it, the samsung was completely assembled and the ipad just had a few compenents so it was easier, the samsung was already opened up before so removing the oled was easy. Cleaned them: the samsung didnt really need that much cleaning but the ipad.. oh boy was it something. i would confedently assume it was owned by an ipad kid since it was filled with food crumbs, also, i might or might not broken a few components while removing them, not that i will need them but yea.. Snipped and sanded down the samsung bracket: the samsung tablet display sits behind a plastic bracket that also has all the components in the back of it, this part is very crucial since it holds everything together, so i have to snip it and cut it perfectly until it fits in the ipad. This is about all i did today, not much but tommorow i will borrow a dremel from a friend a use it until then i’ll post more updates, stay goated gng✌"}]...
|
Title: iTab: a 2013 iPad with Android and OLED by Title: iTab: a 2013 iPad with Android and OLED by @water | Stardance
Devlogs:
iTab DEVLOG-1!!! Today we lit🔥🔥🔥 i’ve done multiple improvements to this project!: Disassembled both tablets: ik, boring but figured i’d mention it, the samsung was completely assembled and the ipad just had a few compenents so it was easier, the samsung was already opened up before so removing the oled was easy. Cleaned them: the samsung didnt really need that much cleaning but the ipad.. oh boy was it something. i would confedently assume it was owned by an ipad kid since it was filled with food crumbs, also, i might or might not broken a few components while removing them, not that i will need them but yea.. Snipped and sanded down the samsung bracket: the samsung tablet display sits behind a plastic bracket that also has all the components in the back of it, this part is very crucial since it holds everything together, so i have to snip it and cut it perfectly until it fits in the ipad. This is about all i did today, not much but tommorow i will borrow a dremel from a friend a use it until then i’ll post more updates, stay goated gng✌
iTab DEVLOG-2!!!: i haven’t done much today, but still some progress nonetheless! Cut more off the display bracket: at the bottom where the speakers are and on top where the buttons are, i cut more parts of the bracket, the cuts were rough since my only dremel started smoking the moment i turned it on so i had to only use pliers>:( Dremeled off the iPad housing: I removed unnecessary aluminium especially off the lte antenna (which was blocking most of the upper bracket) not much has been done today since i was tired but i’ll lock in and hopefully finish ts by the end of the week!! until then stay goated gng✌
iTab DEVLOG-3!!!: A few important things were done today!! i’m very happy with how quickly this is coming together🥹 Made the battery pack!: yes you read that right, “made”, you see, the original battery was bloated and since it was old and expensive, i made my own , i pulled 2 almost identical li-ion cells from phones (4000mAh each) and wired them to the original bms which wires them in parallel, i then just wrapped them around electrical tape and now have a cool little 8000mAh 3.85v cell! Removed more off the bracket: Now the boring stuff, like the last 2 devlogs, i dremeled the bracket and the housing properly this time since i got an actual good dremel (borrowed but still , sanded the part that needs sanding more, and yeah now it fits perfectly!, the only thing remaining is making screw holes and i should move on to the motherboard and charging port and buttons…etc Aaaand yea thats all for today, tomorrow will be even more lit than today, until then stay goated gng✌
iTab DEVLOG-4!! i haven’t posted anything in the past 3 days bc i was cooking up smth big , ive done so much to this project so far and i can say we’re halfway through it, here is everything i’ve done in the past 3 days: Made screw holes for the bracket to screw to the iPad frame: Since the tablet’s bracket was just sitting lose on the frame, i made some screw holes in frame out of (i kid you not) sim cards plastic placeholders , sounds weird ik but it works and sticks to metal very well since its just thick plastic, plus it elevated the bracket just enough to make it sit flush with the frame Added USB-C!!! Now THIS is a big one, i sourced a USB-C breakout board from a local repair shop, i was planning on going through the lengthy process of using a full Type-C port and adding resistors to the CC pins and shi, but i got lucky , i used superglue and hot glue to secure it in place and it suprisingly looks very good!! Wired all the I/O Power button, vol buttons, headphone jack, and Type-C connector, i soldered wires directly to them and routed them with electical tape, i also commoned all ground to the metal chasis Made a removable connector for the I/O!! in addition to wiring and routing I/O, i couldn’t just solder everything together since i know DAMN WELL i something will break and i’d find myself reslodering wires same (if not less) thickness than my hair, so i used a ZIF connector (or whatever it’s called) i pulled from a dead laptop to connect both sides, and i had to solder BOTH sides with of the connector with horrible lighting AND with a $3 iron i had since 2023 AND WITHOUT a microscope , let that sink in.. other than that i made a few minor improvments like adding a placebo camera lens for looks (since i’m NOT rewiring the camera from center to side without a microscope) and also made sure everything is routed neatly, i used specifically wires from old headphones since they were thin and insulated. Anyways that’s all for today!, i’d say we’re 50-55% through the project, byt he end of this week we will hopefully have finished it, i’ve also noticed this project going viral! thanks for all the support, and until next time, stay goated gng✌
iTab DEVLOG-5!! not a revolutionary devlog like the last one byt still, many things have changed, and this time, i’ve finally got to test it out and share the results!!🥹 if you’re not intrested in the latest changes i’ve done, you can skip to the end and check out the test results! Improved the battery pack! if you seen one of my earlier devolgs, you’d probably know that i made a battery pack for the tablet with 2 phone batteries, that idea is fine, it’s cheap and offers an even higher capacity than the original battery! but who couldve guessed that wrapping an already thick part in 5 rounds in electrical tape would make it any thicker , i just unwrapped it and made sure its insulated. Reverted some changes!! i didnt really need the headphone jack nor the data pins so i removed them, i also got rid of the ZIF connector and just soldered the wires directly!, i know its kinda lazy and defeats alot of features but i genuinely dont need these features since all i’m using the tablet for is watching and studying, and they made the assembly so much harder so they had to go unfourtunately Redone all the button logic! Samsung’s a bitch they made the power button completely insulated from ground, meaning instead of needing to only run 3 wires for button signals ans grouping the ground, i had to run an extra wire JUST for the power button , while i was at it, i also mapped out what each pin connected to Tested out!!!! since the start of the project, i haven’t been able to test it out, but now, i am!!, here are what works and what doesn’t!: What works: Power button Volume up button Bluetooth Mic Basic functions Battery Charging (ignore the temp error the battery connector is just loose) What doesn’t work: Wi-Fi (Signal very low, since the tablet used to rely on it’s metal frame as an antenna) Volume down button (Surprisingly) USB-C Data (Didn’t wire them on purpose) Speakers (Very surprisingly, i thought i wired them perfectly The conclusion i’d say we’re ~65% into the project, i still need to fix all the problems i mentioned, and make everything easily accessible before we move to the last (and frankly the hardest) step, which is delaminating and replacing the OLED’s glass, and until then, stay goated gng✌...
|
1792918080.2950504
|
1793782080.2950504
|
|
67104
|
iWebOS-iPhone in the Web by @GoodAron | Stardance
|
GoodAron
|
1
|
13.0
|
0
|
[{"ship_num": 1, "ts": "2026-1 [{"ship_num": 1, "ts": "2026-10-07T19:12:50Z", "hours": 13.0, "multiplier": null, "payout": 0, "blessing": "neutral", "text": "iPhone in the web-an OS that looks as close to an iPhone as i could get it, keep in mind i don’t have an iPhone", "repo": "https://github.com/GoodShabbos/artists-os", "demo": "https://goodshabbos.github.io/iwebos/"}]...
|
[{"post_id": 77659, "ts": "202 [{"post_id": 77659, "ts": "2026-10-07T19:11:11Z", "hours": 6.3642, "text": "ok i think i am going to ship, so this will be a final devlog, but guys if you see any way i can improve, please comment and i will fix! #iPhoneInTheWeb📱📱👍 for all you nerds you want to know what i did for these six hours i made the lock screen unlock normaly, i added three more backgrounds and a button to rotate backgrounds, a clock in the top right corner, the credits panel which is terrible, and in the folder app a new add note button which took donkeys to make, and now the light mode is so much better"}, {"post_id": 76995, "ts": "2026-10-05T09:20:26Z", "hours": 1.1175, "text": "okay guys the iPhone in the web is looking quite sick if i say so my self i have fixed a load of bugs: when i made the iphone animation of tabs springing up if you know what i mean, then also added an animation for the tabs at the bottom, also now the calculator actually works, i have to see now about getting a browser in there let me take a look until next time! #️⃣📱IphoneInTheWeb"}, {"post_id": 76136, "ts": "2026-10-02T13:41:16Z", "hours": 0.3428, "text": "iPhone in the web i added some fake status (battery and wifi and signal), and the island which i only now noticed iphones 📱have, and the search which i haven’t wired up yet, also brightness slider which uses black overlay to not glitch out the liquid glass"}, {"post_id": 76128, "ts": "2026-10-02T13:03:06Z", "hours": 0.6681, "text": "made it look so much more like an iphone the iphone in the web is looking good although i still need some more bits and pieces to finish with, not so far off!"}, {"post_id": 76119, "ts": "2026-10-02T11:22:30Z", "hours": 0.3517, "text": "iPhone in the Web, iWebOS update: ok i have cleaned up the folder explorer, notepad (uses md) all apps at the moment are re-using the same components so once i had the file explorer the rest was easy"}, {"post_id": 76112, "ts": "2026-10-02T10:41:33Z", "hours": 0.3775, "text": "ok guys the iPhone in the web now has two movable tabs-background chooser and credits (doesn’t do anything, background chooser somehow works) am not fully happy with the design so might spend more time over it, let me see 👍👍"}, {"post_id": 76104, "ts": "2026-10-02T10:13:51Z", "hours": 0.4056, "text": "ok we have the start of an OS over here still no backend or wiring but the project is looking nice and neat-these tabs are just static and dont do anything, I copied and pasted the logos into the folder, now we have to see about some sort of file explorer next"}, {"post_id": 76087, "ts": "2026-10-02T09:37:58Z", "hours": 0.3331, "text": "ok i have finalised the homescreen with some touches, now with 7 rotating backgrounds, now i need to make the os! also the login screen although it doesnt do anything yet, still have to wire it up"}, {"post_id": 75970, "ts": "2026-10-01T19:42:03Z", "hours": 0.5269, "text": "homescreen/landing page is up-it looks 👌i’ll admit that the background is an ai made JPG file but it looks good for my iPhone in the Web-iWebOS sadly i have spend ages making this (3 hours) and i have no OS and no backend but i’ll get there hopefuly"}, {"post_id": 75959, "ts": "2026-10-01T18:51:26Z", "hours": 0.325, "text": "this clock is sick, managed to get hold of the sf pro font, used liquid glass components from liqui glass and some other random repo, image in background is ai generated but is still insane"}, {"post_id": 75886, "ts": "2026-10-01T14:24:11Z", "hours": 1.2564, "text": "Changed the project: ipone OS, in the web! (only screensaver as it is, plis heart 📱👍)"}, {"post_id": 75683, "ts": "2026-09-30T19:04:57Z", "hours": 0.3214, "text": "basic frontend is up! worked on it now, ai scaffolded and i went over with what the guide said"}, {"post_id": 75677, "ts": "2026-09-30T18:38:31Z", "hours": 0.2764, "text": "Scaffolded whole project based, backend only, created github repo, going to sort out hosting and check out the guide for the frontend-going to make a canvas OS for artists!"}]...
|
Title: iWebOS-iPhone in the Web by @GoodAron | Sta Title: iWebOS-iPhone in the Web by @GoodAron | Stardance
Devlogs:
Scaffolded whole project based, backend only, created github repo, going to sort out hosting and check out the guide for the frontend-going to make a canvas OS for artists!
basic frontend is up! worked on it now, ai scaffolded and i went over with what the guide said
Changed the project: ipone OS, in the web! (only screensaver as it is, plis heart 📱👍)
this clock is sick, managed to get hold of the sf pro font, used liquid glass components from liqui glass and some other random repo, image in background is ai generated but is still insane
homescreen/landing page is up-it looks 👌i’ll admit that the background is an ai made JPG file but it looks good for my iPhone in the Web-iWebOS sadly i have spend ages making this (3 hours) and i have no OS and no backend but i’ll get there hopefuly
ok i have finalised the homescreen with some touches, now with 7 rotating backgrounds, now i need to make the os! also the login screen although it doesnt do anything yet, still have to wire it up
ok we have the start of an OS over here still no backend or wiring but the project is looking nice and neat-these tabs are just static and dont do anything, I copied and pasted the logos into the folder, now we have to see about some sort of file explorer next
ok guys the iPhone in the web now has two movable tabs-background chooser and credits (doesn’t do anything, background chooser somehow works) am not fully happy with the design so might spend more time over it, let me see 👍👍
iPhone in the Web, iWebOS update: ok i have cleaned up the folder explorer, notepad (uses md) all apps at the moment are re-using the same components so once i had the file explorer the rest was easy
made it look so much more like an iphone the iphone in the web is looking good although i still need some more bits and pieces to finish with, not so far off!
iPhone in the web i added some fake status (battery and wifi and signal), and the island which i only now noticed iphones 📱have, and the search which i haven’t wired up yet, also brightness slider which uses black overlay to not glitch out the liquid glass
okay guys the iPhone in the web is looking quite sick if i say so my self i have fixed a load of bugs: when i made the iphone animation of tabs springing up if you know what i mean, then also added an animation for the tabs at the bottom, also now the calculator actually works, i have to see now about getting a browser in there let me take a look until next time! #️⃣📱IphoneInTheWeb
ok i think i am going to ship, so this will be a final devlog, but guys if you see any way i can improve, please comment and i will fix! #iPhoneInTheWeb📱📱👍 for all you nerds you want to know what i did for these six hours i made the lock screen unlock normaly, i added three more backgrounds and a button to rotate backgrounds, a clock in the top right corner, the credits panel which is terrible, and in the folder app a new add note button which took donkeys to make, and now the light mode is so much better
Ships:
iPhone in the web-an OS that looks as close to an iPhone as i could get it, keep in mind i don’t have an iPhone...
|
1793591885.7004547
|
1793678285.7004547
|
|
19778
|
iWrite4.me by @kamoldme | Stardance
|
kamoldme
|
0
|
14.13
|
0
|
[]
|
[{"post_id": 9956, "ts": "2026 [{"post_id": 9956, "ts": "2026-06-16T14:15:48Z", "hours": 14.1306, "text": "🕹️ iWrite4.me, the distraction free writing app I’ve been building. #Quick_update The idea is simple. You open it, a timer starts, and you write. No AI finishing your sentences for you, no tabs to slip away to. There’s even a Dangerous mode where if you stop typing or leave the page, your work gets deleted. It sounds harsh, but it’s the feature people end up loving, because it actually forces you to finish the thought or the essay you were procrastinating for almost a month. Where it stands right now: Around 500+ users signed up 110,000+ words written on the platform 100+ hours of focused writing logged What I’ve shipped recently: • Focus sounds built in (brown noise, 40 Hz gamma, lofi), plus you can paste any YouTube link to play in the background • Daily streaks, XP and levels, leaderboards and achievements to keep the habit going • A Community tab where people post their stories and essays and get some feedback/comments • Documents organized in folders, PDF export, public profiles, Google login • Stripe subscriptions are live It’s running in production at iwrite4.me. Next I want to grow the community side and make the writing stats feel more rewarding. Slowly turning a small side project into something people open every day. will be glad for your support with “likes”😅"}]...
|
Title: iWrite4.me by @kamoldme | Stardance
Devlog Title: iWrite4.me by @kamoldme | Stardance
Devlogs:
🕹️ iWrite4.me, the distraction free writing app I’ve been building. #Quick_update The idea is simple. You open it, a timer starts, and you write. No AI finishing your sentences for you, no tabs to slip away to. There’s even a Dangerous mode where if you stop typing or leave the page, your work gets deleted. It sounds harsh, but it’s the feature people end up loving, because it actually forces you to finish the thought or the essay you were procrastinating for almost a month. Where it stands right now: Around 500+ users signed up 110,000+ words written on the platform 100+ hours of focused writing logged What I’ve shipped recently: • Focus sounds built in (brown noise, 40 Hz gamma, lofi), plus you can paste any YouTube link to play in the background • Daily streaks, XP and levels, leaderboards and achievements to keep the habit going • A Community tab where people post their stories and essays and get some feedback/comments • Documents organized in folders, PDF export, public profiles, Google login • Stripe subscriptions are live It’s running in production at iwrite4.me. Next I want to grow the community side and make the writing stats feel more rewarding. Slowly turning a small side project into something people open every day. will be glad for your support with “likes”😅...
|
1793393306.2535346
|
1793998106.2535346
|
|
65739
|
ia by @jrosario676 | Stardance
|
jrosario676
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: ia by @jrosario676 | Stardance
|
1793091037.8420389
|
1793695837.8420389
|
|
29917
|
icanhelp by @TheShovel | Stardance
|
TheShovel
|
1
|
54.0
|
1020
|
[{"ship_num": 1, "ts": "2026-0 [{"ship_num": 1, "ts": "2026-07-22T09:04:44Z", "hours": 54.0, "multiplier": 19.73, "payout": 1020, "blessing": "neutral", "text": "I decided to make an AI chatbot, that runs locally, without burning your computer. It is designed around using a 500mb, 0.8B model and to work even on weak hardware. Compatibility and actually giving you a result eventually was the priority, not speed! I also made a YouTube video about it if you wanna check that out :3 https://www.youtube.com/watch?v=GhmLUcdesqw", "repo": "https://github.com/TheShovel/icanhelp", "demo": "https://theshovel.rocks/icanhelp/"}]...
|
[{"post_id": 32596, "ts": "202 [{"post_id": 32596, "ts": "2026-07-21T21:27:35Z", "hours": 0.2581, "text": "just a small patch related to memory usage, nothing more :) though, the video is ready! im just waiting for it to get approved for the hackclub video thing and then im gonna post it! Heres the thumbnail"}, {"post_id": 31199, "ts": "2026-07-19T12:23:08Z", "hours": 5.2222, "text": "i think i am finally done! I have optimized, refined stuff and tested a bunch of random things on the lowest model possible, that still has some kind of a intelligence. Nothing really too interesting to talk about, but i DID get it working on my 7gb of ram old laptop! This is the first local chat app that actually works on that thing, so thats pretty damn cool me thinks. Working on a youtube video right now, im proud of the project and i want people to find it and contribute to it. In the meantime, until i finish that video, here is icanhelp writing a document about the importance of vitamin C!"}, {"post_id": 29946, "ts": "2026-07-17T04:07:10Z", "hours": 9.3467, "text": "finally a devlog where not everything was a colossal failure! As i said in my previous dev log, whats left to do is tools and refinement, and thats exactly what i did. The search tool has been updated to let the llm browse websites and their contents, which improved the quality a lot. also, i have implemented an addon system that lets you install external smaller llms to do tasks that are then outputed to the bigger llm. llmception. decided to get rid of the rag since with the improved web search its no longer needed. Besides this, all i did was little tweaks, refinement refinement testing refinement. As one smart video once said: “solution problem, problem solution problem.”"}, {"post_id": 28767, "ts": "2026-07-15T07:30:06Z", "hours": 12.7231, "text": "well well well. This project has been the most frustrating back and forth of my life. It felt like 3 steps forward, 2.5 steps back, CONSTANTLY. Well, after hours and hours of testing, refinement and research, i have found the solution… I have decided to change my approach in how i handle responses. Now, instead of having 1 model that does everything, we have 2 super tiny ones. The flow is kinda like: user types prompt -> small llm analyzes prompt -> small llm returns the instructions for how to solve it using the tools it has -> if it decides that the complexity is too high, then it lets the larger llm continue -> system does knowladgebase + web search -> the chosen llm presents the data to you. This is REALLY fast as you can see, and quite reliable as well. Right now the answer quality is not exceptional, but thats just about refinement and gradual improvements. So hopefully in the next devlog i have some really cool stuff to show you, like graphs, visual explanations, letting the user choose a website to summarize and so on. Along side working on the app, i worked on the website for it! Its still a work in progress but im proud of how it turned out :D https://theshovel.rocks/icanhelp/ ok well, anyway, thats it for now. see you next time in “im going crazy because lobotomy ai doesnt want to follow my instructions”"}, {"post_id": 27563, "ts": "2026-07-13T08:01:26Z", "hours": 8.0653, "text": "well, today was a disaster. 8 ours that didnt really account for much to be honest :( I have tried to move from qwen to minicpm 1b, a model that ALLEGEDLY is a lot better at using tools to gather data rather than using its internal baked in weights. Well, apparently thats not true. I have spent hours making the tools, adapting the outputs and everything for that model, just to then realized that “wow people are wrong and im so tired of ai bros hyping stuff up for nothing!”. So, in the end, i reverted to the qwen models, since they are decent enough for this. Even the 0.8b model gives better results than minicpm… Well, at least thats not all i did. After reverting to a older commit, i have re implemented all the improvements that dont have to do with minicpm. You can now see the sources of where the llm fetched data from under the response bubble, and you can click them to open up the pages and double check the information and stuff. Also, i have worked on large file handling. Previously, for large files it would try to read the entire thing at once. That is insanely inefficient for large files, so i have added a new tool that lets it read from line x to line y. Like that it can only retrieve relevant stuff and not flood its context. I have ALSO worked on compatibility. Now you can change between cpu and gpu inference. If you have under 6gb of vram, the gpu option is grayed out. You might say “thats a little overkill”, but you have to take into account that context also eats up vram, and quite a lot as well. So now you can use the cpu, which uses your system ram, to reliably have long conversations. Sadly in the demo it ended up crashing the app and freezing the process with a node api error, but after i killed the process and reopened the app, the cpu inference worked fine. Thats a small oversight i didnt take into account where llamacpp is still doing stuff while im trying to kill the node process. Well, thats it! See you next time in “trying to make lobotomy ai useful so people stop using giant datacenters to ask about how long to boil and egg for”"}, {"post_id": 26597, "ts": "2026-07-11T14:51:17Z", "hours": 6.4203, "text": "well, i am going to sleep for a few days. The last 24 hours have been tough. First of all, i started getting harassed for making an ai project in my friend groups, which just broke my heart. I got actively attacked and called names for just wanting to make something cool. The point of this project is to make a completely local, light and fast ai assistant for your pc. I dont like the big corporations that do harm economically and environmentally. That is exactly why im making this. With that out of the way, i have spent way longer than 6 hours and a half since the last devlog, bug a lot of that time was not coding, it was data stuff. I have started to build a knowledge base by asking local larger ai that run slower, to write information and documentation for different topics (linux, programming, science, health etc). Right now i have 10 categories with a total of 463 files. Then, i run an embedding model to turn those into searchable vectors that the ai can use during conversation. In the video bellow, i asked it how bash commands work in linux, and it found all the knowladge for that, and started telling me about it. Normally, the quality of the response is pretty short, vague and not very precise, since the model is so small, but now, it actually gives some usable output. I have also worked a lot on making this as easy to install as possible. I have made a install script that covers all dependencies you need, registers the app on your system, and also lets you choose what knowledge categories you want to be included in the embedding process, which also happens locally during install. Yes that means all that knowledge if free and open source in the repository. If you want, you could even make your own categories, or improve existing ones! To make this easier to use, i have also worked on the UI. Now theres a built in light and dark mode that you can use, along side whatever themes you ask the assistant to make for you. You can also now open all of the apps data folders. You can change whatever you want in there, and even customize your buddy! Would be cool to see some user made designs with like miku or something. That would be so cool :3 Sadly, in the process, i also broked the vision model stuff and i just realized that when i recorded the video for this devlog. I am too tired after doing data training for an entire day to debug and fix that right now. Also pretty demotivated after the stuff i had to endure lately for making projects based on ai. I want to remind you that most of this project is made by me, and like my other project for stardance, Irongingot, i will make a graph with all of the files that were touched by ai, how many lines it wrote and what % is ai in the readme of the repository, using my opencode history on this project. I am a human being making stuff for human beings. Anyway, pretty grim devlog i know. Hopefully, the next one will be happier and i can show some more cool stuff i have planned, like real time interaction with your desktop and select text and ask and so on. Thank you to everyone who has been kind to me on here <3"}, {"post_id": 25787, "ts": "2026-07-10T07:37:39Z", "hours": 5.8194, "text": "well well well… turns out cloud AIs are expensive and dont work offline. Also not very accessible for everyone to use!!! So, i have decided to make all of this run LOCALLY!!! Thats right, instead of making you put in an api key to whatever provider you wanna use, you instead get to choose from a list of qwen 3.5 models. It auto detects your hardware, and displays all of the models that work on your system. Now, obviously, this is less high quality than the cloud models. In the previous devlogs i used deepseek v4 flash from opencode, and it was way more competent. Here, i used the 2B version of qwen 3.5. With that said, even the 0.8B model is still fairly capable, considering that deepseek v4 flash is over 200b parameters. I worked a lot on optimizing this. I added chat history compression, so instead of it remembering everything in the conversation precisely, it makes a summary within a token budget instead, to keep system requirements low even for the bigger models. More context tokens = more ram, and who the hell has ram nowadays, am i right gamers. Along with that i made a bunch of small tool and ui improvements. Still a work in progress, but at least the UI is not vibe coded slop for the most part anymore. Next up im going to try and work on a theory that i have. What if i give the small 0.8B model, access to a dataset, and a search engine… Could i possibly make this super tiny model really smart? I dont know. Saw a youtube video about it but it was just a theory, not applied practically. But since im here might as well try it out. Oh also i renamed the little guy to Icanhelpy because i think its very cute"}, {"post_id": 24810, "ts": "2026-07-08T16:43:53Z", "hours": 4.9122, "text": "I got a lot done. Added OCR with tesseract and then i realized it sucked so i switched to BLIP. Runs in a child process so it doesnt freeze the app. The buddy icon animates now based on whats happening, pulses tilts shakes spins depending on state. Also i drew some nice little assets for it :3 Moved the effort dropdown and model selector to the top bar. Settings has sub menus, theme panel lets you browse saved themes and apply or delete them. The AI can change css now through a set_theme tool. Colors gradients images whatever. It can save named themes and they persist across restarts in the encrypted config. Had to consolidate the theme tools cuz the llm request got too big. Screenshots took forever. desktopCapturer asked for permission, imagemagick import has broken syntax, mss gave black screens on wayland, grim doesnt work on kde. Finally spectacle works. Falls back through a chain of tools and tries to auto install imagemagick with a sudo popup if nothing is found. The attach button shows a file and screenshot popup, hides after picking, comes back when you send or remove. Attachment label is a clickable pill that opens files. Fixed the llm re analyzing already attached files. Fixed bubble overflow on long lines. Added chat history, cancel stream, sudo password popup, markdown, web search, file tools, right click menu, window position persistence, burger menu navigation blah blah blah this is getting soooo long bro i dont wanna type no more and a readme. Still need to add proper security beyond the sudo prompt and encrypted config. Gonna work on that next maybe, but thats pretty boring stuff so idk… also sorry for the video quality i just realized my settings were wrong right as i uploaded this. woops"}, {"post_id": 20869, "ts": "2026-07-02T16:18:14Z", "hours": 1.6497, "text": "I have started work on this thing! So far, I have implemented some placeholder ui, a way for you to register your LLM provider, cloud or local, and some basic tools! Copy pasted a lot of code from some of my other projects, so this didnt take too long to do. The thing that took me the longest was to find a method to detect when a console command that the llm runs would return a sudo prompt, so i can prompt the user through the UI to enter their password. So far, thats also the only security thing i have in place. Will work on more of that. The UI is very ugly and clunky right now because its just some css i vibe coded. Trust me i know its ugly ill work on making it better myself once i get everything backend related done."}]...
|
Title: icanhelp by @TheShovel | Stardance
Devlogs Title: icanhelp by @TheShovel | Stardance
Devlogs:
I have started work on this thing! So far, I have implemented some placeholder ui, a way for you to register your LLM provider, cloud or local, and some basic tools! Copy pasted a lot of code from some of my other projects, so this didnt take too long to do. The thing that took me the longest was to find a method to detect when a console command that the llm runs would return a sudo prompt, so i can prompt the user through the UI to enter their password. So far, thats also the only security thing i have in place. Will work on more of that. The UI is very ugly and clunky right now because its just some css i vibe coded. Trust me i know its ugly ill work on making it better myself once i get everything backend related done.
I got a lot done. Added OCR with tesseract and then i realized it sucked so i switched to BLIP. Runs in a child process so it doesnt freeze the app. The buddy icon animates now based on whats happening, pulses tilts shakes spins depending on state. Also i drew some nice little assets for it :3 Moved the effort dropdown and model selector to the top bar. Settings has sub menus, theme panel lets you browse saved themes and apply or delete them. The AI can change css now through a set_theme tool. Colors gradients images whatever. It can save named themes and they persist across restarts in the encrypted config. Had to consolidate the theme tools cuz the llm request got too big. Screenshots took forever. desktopCapturer asked for permission, imagemagick import has broken syntax, mss gave black screens on wayland, grim doesnt work on kde. Finally spectacle works. Falls back through a chain of tools and tries to auto install imagemagick with a sudo popup if nothing is found. The attach button shows a file and screenshot popup, hides after picking, comes back when you send or remove. Attachment label is a clickable pill that opens files. Fixed the llm re analyzing already attached files. Fixed bubble overflow on long lines. Added chat history, cancel stream, sudo password popup, markdown, web search, file tools, right click menu, window position persistence, burger menu navigation blah blah blah this is getting soooo long bro i dont wanna type no more and a readme. Still need to add proper security beyond the sudo prompt and encrypted config. Gonna work on that next maybe, but thats pretty boring stuff so idk… also sorry for the video quality i just realized my settings were wrong right as i uploaded this. woops
well well well… turns out cloud AIs are expensive and dont work offline. Also not very accessible for everyone to use!!! So, i have decided to make all of this run LOCALLY!!! Thats right, instead of making you put in an api key to whatever provider you wanna use, you instead get to choose from a list of qwen 3.5 models. It auto detects your hardware, and displays all of the models that work on your system. Now, obviously, this is less high quality than the cloud models. In the previous devlogs i used deepseek v4 flash from opencode, and it was way more competent. Here, i used the 2B version of qwen 3.5. With that said, even the 0.8B model is still fairly capable, considering that deepseek v4 flash is over 200b parameters. I worked a lot on optimizing this. I added chat history compression, so instead of it remembering everything in the conversation precisely, it makes a summary within a token budget instead, to keep system requirements low even for the bigger models. More context tokens = more ram, and who the hell has ram nowadays, am i right gamers. Along with that i made a bunch of small tool and ui improvements. Still a work in progress, but at least the UI is not vibe coded slop for the most part anymore. Next up im going to try and work on a theory that i have. What if i give the small 0.8B model, access to a dataset, and a search engine… Could i possibly make this super tiny model really smart? I dont know. Saw a youtube video about it but it was just a theory, not applied practically. But since im here might as well try it out. Oh also i renamed the little guy to Icanhelpy because i think its very cute
well, i am going to sleep for a few days. The last 24 hours have been tough. First of all, i started getting harassed for making an ai project in my friend groups, which just broke my heart. I got actively attacked and called names for just wanting to make something cool. The point of this project is to make a completely local, light and fast ai assistant for your pc. I dont like the big corporations that do harm economically and environmentally. That is exactly why im making this. With that out of the way, i have spent way longer than 6 hours and a half since the last devlog, bug a lot of that time was not coding, it was data stuff. I have started to build a knowledge base by asking local larger ai that run slower, to write information and documentation for different topics (linux, programming, science, health etc). Right now i have 10 categories with a total of 463 files. Then, i run an embedding model to turn those into searchable vectors that the ai can use during conversation. In the video bellow, i asked it how bash commands work in linux, and it found all the knowladge for that, and started telling me about it. Normally, the quality of the response is pretty short, vague and not very precise, since the model is so small, but now, it actually gives some usable output. I have also worked a lot on making this as easy to install as possible. I have made a install script that covers all dependencies you need, registers the app on your system, and also lets you choose what knowledge categories you want to be included in the embedding process, which also happens locally during install. Yes that means all that knowledge if free and open source in the repository. If you want, you could even make your own categories, or improve existing ones! To make this easier to use, i have also worked on the UI. Now theres a built in light and dark mode that you can use, along side whatever themes you ask the assistant to make for you. You can also now open all of the apps data folders. You can change whatever you want in there, and even customize your buddy! Would be cool to see some user made designs with like miku or something. That would be so cool :3 Sadly, in the process, i also broked the vision model stuff and i just realized that when i recorded the video for this devlog. I am too tired after doing data training for an entire day to debug and fix that right now. Also pretty demotivated after the stuff i had to endure lately for making projects based on ai. I want to remind you that most of this project is made by me, and like my other project for stardance, Irongingot, i will make a graph with all of the files that were touched by ai, how many lines it wrote and what % is ai in the readme of the repository, using my opencode history on this project. I am a human being making stuff for human beings. Anyway, pretty grim devlog i know. Hopefully, the next one will be happier and i can show some more cool stuff i have planned, like real time interaction with your desktop and select text and ask and so on. Thank you to everyone who has been kind to me on here <3
well, today was a disaster. 8 ours that didnt really account for much to be honest :( I have tried to move from qwen to minicpm 1b, a model that ALLEGEDLY is a lot better at using tools to gather data rather than using its internal baked in weights. Well, apparently thats not true. I have spent hours making the tools, adapting the outputs and everything for that model, just to then realized that “wow people are wrong and im so tired of ai bros hyping stuff up for nothing!”. So, in the end, i reverted to the qwen models, since they are decent enough for this. Even the 0.8b model gives better results than minicpm… Well, at least thats not all i did. After reverting to a older commit, i have re implemented all the improvements that dont have to do with minicpm. You can now see the sources of where the llm fetched data from under the response bubble, and you can click them to open up the pages and double check the information and stuff. Also, i have worked on large file handling. Previously, for large files it would try to read the entire thing at once. That is insanely inefficient for large files, so i have added a new tool that lets it read from line x to line y. Like that it can only retrieve relevant stuff and not flood its context. I have ALSO worked on compatibility. Now you can change between cpu and gpu inference. If you have under 6gb of vram, the gpu option is grayed out. You might say “thats a little overkill”, but you have to take into account that context also eats up vram, and quite a lot as well. So now you can use the cpu, which uses your system ram, to reliably have long conversations. Sadly in the demo it ended up crashing the app and freezing the process with a node api error, but after i killed the process and reopened the app, the cpu inference worked fine. Thats a small oversight i didnt take into account where llamacpp is still doing stuff while im trying to kill the node process. Well, thats it! See you next time in “trying to make lobotomy ai useful so people stop using giant datacenters to ask about how long to boil and egg for”
well well well. This project has been the most frustrating back and forth of my life. It felt like 3 steps forward, 2.5 steps back, CONSTANTLY. Well, after hours and hours of testing, refinement and research, i have found the solution… I have decided to change my approach in how i handle responses. Now, instead of having 1 model that does everything, we have 2 super tiny ones. The flow is kinda like: user types prompt -> small llm analyzes prompt -> small llm returns the instructions for how to solve it using the tools it has -> if it decides that the complexity is too high, then it lets the larger llm continue -> system does knowladgebase + web search -> the chosen llm presents the data to you. This is REALLY fast as you can see, and quite reliable as well. Right now the answer quality is not exceptional, but thats just about refinement and gradual improvements. So hopefully in the next devlog i have some really cool stuff to show you, like graphs, visual explanations, letting the user choose a website to summarize and so on. Along side working on the app, i worked on the website for it! Its still a work in progress but im proud of how it turned out :D https://theshovel.rocks/icanhelp/ ok well, anyway, thats it for now. see you next time in “im going crazy because lobotomy ai doesnt want to follow my instructions”
finally a devlog where not everything was a colossal failure! As i said in my previous dev log, whats left to do is tools and refinement, and thats exactly what i did. The search tool has been updated to let the llm browse websites and their contents, which improved the quality a lot. also, i have implemented an addon system that lets you install external smaller llms to do tasks that are then outputed to the bigger llm. llmception. decided to get rid of the rag since with the improved web search its no longer needed. Besides this, all i did was little tweaks, refinement refinement testing refinement. As one smart video once said: “solution problem, problem solution problem.”
i think i am finally done! I have optimized, refined stuff and tested a bunch of random things on the lowest model possible, that still has some kind of a intelligence. Nothing really too interesting to talk about, but i DID get it working on my 7gb of ram old laptop! This is the first local chat app that actually works on that thing, so thats pretty damn cool me thinks. Working on a youtube video right now, im proud of the project and i want people to find it and contribute to it. In the meantime, until i finish that video, here is icanhelp writing a document about the importance of vitamin C!
just a small patch related to memory usage, nothing more :) though, the video is ready! im just waiting for it to get approved for the hackclub video thing and then im gonna post it! Heres the thumbnail
Ships:
I decided to make an AI chatbot, that runs locally, without burning your computer. It is designed around using a 500mb, 0.8B model and to work even on weak hardware. Compatibility and actually giving you a result eventually was the priority, not speed! I also made a YouTube video about it if you wanna check that out :3 https://www.youtube.com/watch?v=GhmLUcdesqw...
|
1793598694.8801358
|
1793685094.8801358
|
|
53854
|
ice cream by @lara_ibrahim | Stardance
|
lara_ibrahim
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: ice cream by @lara_ibrahim | Stardance
|
1793509083.8208854
|
1794113883.8208854
|
|
59343
|
ichi by @rigid_valmeo | Stardance
|
rigid_valmeo
|
1
|
3.0
|
0
|
[{"ship_num": 1, "ts": "2026-0 [{"ship_num": 1, "ts": "2026-09-17T10:00:08Z", "hours": 3.0, "multiplier": null, "payout": 0, "blessing": "neutral", "text": "I make this website and I’m proud of it:3 It was very challenging lol since it was my first time… but tysm for the amazing tutorials nasa stardance!", "repo": "https://github.com/rigidvalmeo03/ichi-website-real", "demo": "https://rigidvalmeo03.github.io/ichi-website-real/"}]...
|
[{"post_id": 69559, "ts": "202 [{"post_id": 69559, "ts": "2026-09-17T09:57:44Z", "hours": 3.0361, "text": "My Website! AAAAAAAA im so happy i made this. It’s my first time:3"}]...
|
Title: ichi by @rigid_valmeo | Stardance
Devlogs: Title: ichi by @rigid_valmeo | Stardance
Devlogs:
My Website! AAAAAAAA im so happy i made this. It’s my first time:3
Ships:
I make this website and I’m proud of it:3 It was very challenging lol since it was my first time… but tysm for the amazing tutorials nasa stardance!...
|
1792856993.0302327
|
1793720993.0302327
|
|
70196
|
ichi other pet by @rigid_valmeo | Stardance
|
rigid_valmeo
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: ichi other pet by @rigid_valmeo | Stardance
|
1792961251.3334916
|
1793825251.3334916
|
|
70187
|
ichi warioware game by @rigid_valmeo | Stardance
|
rigid_valmeo
|
0
|
1.39
|
0
|
[]
|
[{"post_id": 80383, "ts": "202 [{"post_id": 80383, "ts": "2026-10-17T13:57:39Z", "hours": 1.3939, "text": "Making my warioware rn"}]...
|
Title: ichi warioware game by @rigid_valmeo | Star Title: ichi warioware game by @rigid_valmeo | Stardance
Devlogs:
Making my warioware rn...
|
1792961242.3129616
|
1793825242.3129616
|
|
54961
|
icp-scout by @qqxzew | Stardance
|
qqxzew
|
1
|
34.0
|
0
|
[{"ship_num": 1, "ts": "2026-0 [{"ship_num": 1, "ts": "2026-09-07T14:22:09Z", "hours": 34.0, "multiplier": null, "payout": 0, "blessing": "neutral", "text": "What did you make? An assistant for a salesperson, not another lead list. You give it a customer profile, it walks the Czech public registers, and once a week it hands back a handful of companies with something that actually changed this week — each with a full dossier where every claim carries the exact quote it came from and a link to the page it was found on. Built as a take-home assignment for RTsoft (they sell custom shop-floor management software), and it’s live in production, not just running on my machine. What was challenging? Hallucination is the whole risk of this task, not a side concern — an invented fact about a company that a salesperson pastes into an email is worse than no output at all. So the model never gets to assert anything; it proposes a claim + a quote, and my code searches for that exact quote in the page text I archived. First time I ran a verifier that rechecked stored claims against their own archived pages, 100 of 172 supposedly-verified facts failed — all from one code path, the part reading the state register, which had been marking its own claims “verified” without ever passing them through the check. The other hard part: I assumed a good weighted score could rank “why this company.” Two separate experiments — a permutation test and a 1536-dim embedding of company sites — both came back at chance. The site text just doesn’t say who’s buying software. Had to throw the score out and rebuild ranking around what’s actually provable instead. What are you proud of? One line in the schema: claim.snapshot_id NOT NULL REFERENCES snapshot(id). A claim about a company physically cannot be inserted into the database unless it points at a stored page snapshot — “nothing without a source” isn’t a rule I have to remember, it’s a foreign key. Also proud that manual review keeps finding real bugs the statistics don’t: a solar-panel subsidy misfiled as “scaling up production,” a company ranked #1 whose own grant paperwork said they’d already bought the system — both caught by reading five actual cards, after two rigorous experiments said everything was fine. What should people know so they can test it? It only works for Czech companies — the sources are Czech state registers. python -m pipeline.run –check tells you exactly what’s missing before you run anything. The number that matters on every dossier is the last line — “N verified facts · N inferences · N discarded” — click a claim’s source link and check the quote is actually on that page; that’s the whole trust model, so it should be checkable, not taken on faith.", "repo": "https://github.com/qqxzew/icp-scout", "demo": "https://www.icp-scout.fun/"}]...
|
[{"post_id": 63571, "ts": "202 [{"post_id": 63571, "ts": "2026-09-07T14:20:11Z", "hours": 7.4442, "text": "Wired up actual deploy: push to main, and a minute later the demo is running that commit. git archive straight over ssh into a receive script — no registry, no git pull on the server, so the server never holds credentials for this repo. The one thing it refuses to do is kill a run in progress: /api/run spawns the pipeline as a subprocess, and a container restart takes it down with it. Now the script asks first, and if a run is going it stops and exits non-zero — red, on purpose, because a deploy that shipped code but didn’t put it into service has no business looking green.Rebuilding the whole thing from nothing on a clean machine found what months of running it on my own box never did. All five delivered companies 404’d after the first real run — the candidate list gets loaded once at import, a run rewrites the files under it, and the process just kept serving eight-hour-old data until somebody restarted it by hand. The “what’s missing” preflight check was lying in both directions: it called the API key missing on a machine that had it as an env var instead of a .env file, and it reported success on a machine where the industry codebook never downloaded — which doesn’t crash anything, it just prints every industry as a bare five-digit code and calls that fine.README got rewritten to lead with the one sentence the tool actually stands on, plus an ARCHITECTURE.md the code can point to instead of explaining itself in comments, a Czech translation, and an MIT licence with the honest caveat: what’s verified is that a sentence was on a page, not that it’s true."}, {"post_id": 61785, "ts": "2026-09-04T21:17:56Z", "hours": 6.3292, "text": "Built a carousel for the week’s five cards - swipe through them one at a time, or flip to a plain grid"}, {"post_id": 60452, "ts": "2026-09-02T19:46:09Z", "hours": 5.5703, "text": "Ran two experiments trying to find better weights for the outreach-order score — a permutation test on our own signals, and a 1536-dim embedding of company sites — both landed at chance (AUC 0.39, p≈0.8). Turns out public site text just doesn’t say whether a company is buying planning software; that’s a ceiling in the source, not something more features would fix, so scoring got replaced with an order read straight off the client’s own ICP document instead of invented weights. Then reading five actual weekly cards by hand caught two things neither experiment did: a solar-panel subsidy misfiled as “scaling up production,” and the #1 company on the list whose own grant description said “we’re buying a complete information system” — already a lost lead, not a hot one. Fixed the classifier — it was only reading project titles, never descriptions — and one signal class jumped from 214 to 1014 matches, 415 of which used to be pitched as reasons to call. Two rigorous measurements found nothing; two hours of eyeballing cards found the bugs sitting at the top of the list."}, {"post_id": 57638, "ts": "2026-08-29T20:55:41Z", "hours": 2.9286, "text": "The assistant now produces real company cards - every claim on one carries the exact quote it came from and a link to the source it was found in. To make sure that actually held, I wrote a check that re-verifies every stored fact against the archived page it supposedly came from, and ran it over the whole database.100 of 172 “facts” failed. Every single one came from my own code: the part that reads the state company register was writing claims marked “verified” without ever passing them through the verifier. The irony is that the register is the most trustworthy source in the entire project - it was just the only one asserting things instead of proving them. Now 177 of 177 pass.I also cut the weekly run from ~50 minutes to about 2, after finding an official API that lists which companies changed recently - 30 of mine in a week, instead of re-checking all 3,299."}, {"post_id": 55490, "ts": "2026-08-27T11:36:43Z", "hours": 11.6494, "text": "I started working on icp-scout - an AI assistant that helps find potential B2B customers. The idea behind the project is simple: instead of a salesperson spending hours searching for companies, opening different websites, and collecting information manually, the assistant should handle this routine work. The goal is not just to provide a list of companies, but to explain why each company could be a good fit and why now might be the right time to contact them. Right now, I’m working on the basic project architecture and thinking about how to make AI results more reliable. The biggest challenge with systems like this is that AI can confidently generate information that is not true. That’s why it is important that every important statement can be verified and has a clear source. Next, I plan to connect public company data sources, build the first analysis pipeline, and test it with real examples."}]...
|
Title: icp-scout by @qqxzew | Stardance
Devlogs:
Title: icp-scout by @qqxzew | Stardance
Devlogs:
I started working on icp-scout - an AI assistant that helps find potential B2B customers. The idea behind the project is simple: instead of a salesperson spending hours searching for companies, opening different websites, and collecting information manually, the assistant should handle this routine work. The goal is not just to provide a list of companies, but to explain why each company could be a good fit and why now might be the right time to contact them. Right now, I’m working on the basic project architecture and thinking about how to make AI results more reliable. The biggest challenge with systems like this is that AI can confidently generate information that is not true. That’s why it is important that every important statement can be verified and has a clear source. Next, I plan to connect public company data sources, build the first analysis pipeline, and test it with real examples.
The assistant now produces real company cards - every claim on one carries the exact quote it came from and a link to the source it was found in. To make sure that actually held, I wrote a check that re-verifies every stored fact against the archived page it supposedly came from, and ran it over the whole database.100 of 172 “facts” failed. Every single one came from my own code: the part that reads the state company register was writing claims marked “verified” without ever passing them through the verifier. The irony is that the register is the most trustworthy source in the entire project - it was just the only one asserting things instead of proving them. Now 177 of 177 pass.I also cut the weekly run from ~50 minutes to about 2, after finding an official API that lists which companies changed recently - 30 of mine in a week, instead of re-checking all 3,299.
Ran two experiments trying to find better weights for the outreach-order score — a permutation test on our own signals, and a 1536-dim embedding of company sites — both landed at chance (AUC 0.39, p≈0.8). Turns out public site text just doesn’t say whether a company is buying planning software; that’s a ceiling in the source, not something more features would fix, so scoring got replaced with an order read straight off the client’s own ICP document instead of invented weights. Then reading five actual weekly cards by hand caught two things neither experiment did: a solar-panel subsidy misfiled as “scaling up production,” and the #1 company on the list whose own grant description said “we’re buying a complete information system” — already a lost lead, not a hot one. Fixed the classifier — it was only reading project titles, never descriptions — and one signal class jumped from 214 to 1014 matches, 415 of which used to be pitched as reasons to call. Two rigorous measurements found nothing; two hours of eyeballing cards found the bugs sitting at the top of the list.
Built a carousel for the week’s five cards - swipe through them one at a time, or flip to a plain grid
Wired up actual deploy: push to main, and a minute later the demo is running that commit. git archive straight over ssh into a receive script — no registry, no git pull on the server, so the server never holds credentials for this repo. The one thing it refuses to do is kill a run in progress: /api/run spawns the pipeline as a subprocess, and a container restart takes it down with it. Now the script asks first, and if a run is going it stops and exits non-zero — red, on purpose, because a deploy that shipped code but didn’t put it into service has no business looking green.Rebuilding the whole thing from nothing on a clean machine found what months of running it on my own box never did. All five delivered companies 404’d after the first real run — the candidate list gets loaded once at import, a run rewrites the files under it, and the process just kept serving eight-hour-old data until somebody restarted it by hand. The “what’s missing” preflight check was lying in both directions: it called the API key missing on a machine that had it as an env var instead of a .env file, and it reported success on a machine where the industry codebook never downloaded — which doesn’t crash anything, it just prints every industry as a bare five-digit code and calls that fine.README got rewritten to lead with the one sentence the tool actually stands on, plus an ARCHITECTURE.md the code can point to instead of explaining itself in comments, a Czech translation, and an MIT licence with the honest caveat: what’s verified is that a sentence was on a page, not that it’s true.
Ships:
What did you make? An assistant for a salesperson, not another lead list. You give it a customer profile, it walks the Czech public registers, and once a week it hands back a handful of companies with something that actually changed this week — each with a full dossier where every claim carries the exact quote it came from and a link to the page it was found on. Built as a take-home assignment for RTsoft (they sell custom shop-floor management software), and it’s live in production, not just running on my machine. What was challenging? Hallucination is the whole risk of this task, not a side concern — an invented fact about a company that a salesperson pastes into an email is worse than no output at all. So the model never gets to assert anything; it proposes a claim + a quote, and my code searches for that exact quote in the page text I archived. First time I ran a verifier that rechecked stored claims against their own archived pages, 100 of 172 supposedly-verified facts failed — all from one code path, the part reading the state register, which had been marking its own claims “verified” without ever passing them through the check. The other hard part: I assumed a good weighted score could rank “why this company.” Two separate experiments — a permutation test and a 1536-dim embedding of company sites — both came back at chance. The site text just doesn’t say who’s buying software. Had to throw the score out and rebuild ranking around what’s actually provable instead. What are you proud of? One line in the schema: claim.snapshot_id NOT NULL REFERENCES snapshot(id). A claim about a company physically cannot be inserted into the database unless it points at a stored page snapshot — “nothing without a source” isn’t a rule I have to remember, it’s a foreign key. Also proud that manual review keeps finding real bugs the statistics don’t: a solar-panel subsidy misfiled as “scaling up production,” a company ranked #1 whose own grant paperwork said they’d already bought the system — both caught by reading five actual cards, after two rigorous experiments said everything was fine. What should people know so they can test it? It only works for Czech companies — the sources are Czech state registers. python -m pipeline.run –check tells you exactly what’s missing before you run anything. The number that matters on every dossier is the last line — “N verified facts · N inferences · N discarded” — click a claim’s source link and check the quote is actually on that page; that’s the whole trust model, so it should be checkable, not taken on faith....
|
1793670497.4004524
|
1793756897.4004524
|
|
70208
|
id card holder by @killerbelugawhale | Stardance
|
killerbelugawhale
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: id card holder by @killerbelugawhale | Star Title: id card holder by @killerbelugawhale | Stardance...
|
1793606000.4769151
|
1794210800.4769151
|
|
30126
|
ide by @venk | Stardance
|
venk
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: ide by @venk | Stardance
|
1793411837.7487063
|
1794016637.7487063
|
|
8721
|
idea evolution engine by @elvitigalalis | Stardanc idea evolution engine by @elvitigalalis | Stardance...
|
elvitigalalis
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idea evolution engine by @elvitigalalis | S Title: idea evolution engine by @elvitigalalis | Stardance...
|
1793170122.7476475
|
1793774922.7476475
|
|
621
|
idea loading... by @claraaa | Stardance
|
claraaa
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idea loading... by @claraaa | Stardance
|
1793487131.1784842
|
1794091931.1784842
|
|
19488
|
idk :p by @dhanvitha_ | Stardance
|
dhanvitha_
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk :p by @dhanvitha_ | Stardance
|
1793134145.6543007
|
1793738945.6543007
|
|
54693
|
idk by @67 | Stardance
|
67
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @67 | Stardance
|
1793559406.7655478
|
1794164206.7655478
|
|
70125
|
idk by @Ankit_7 | Stardance
|
Ankit_7
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @Ankit_7 | Stardance
|
1792958083.9264371
|
1793822083.9264371
|
|
71576
|
idk by @Casinii | Stardance
|
Casinii
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @Casinii | Stardance
|
1793098025.0607305
|
1793702825.0607305
|
|
56395
|
idk by @Christianlikes67_Bobbpheus | Stardance
|
Christianlikes67_Bobbpheus
|
0
|
0.27
|
0
|
[]
|
[{"post_id": 58675, "ts": "202 [{"post_id": 58675, "ts": "2026-08-31T05:48:46Z", "hours": 0.2678, "text": "yeeeessss"}]...
|
Title: idk by @Christianlikes67_Bobbpheus | Starda Title: idk by @Christianlikes67_Bobbpheus | Stardance
Devlogs:
yeeeessss...
|
1792921789.088816
|
1793785789.088816
|
|
3429
|
idk by @SillySilicon | Stardance
|
SillySilicon
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @SillySilicon | Stardance
|
1793526667.1999173
|
1794131467.1999173
|
|
66831
|
idk by @VirajSingh15 | Stardance
|
VirajSingh15
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @VirajSingh15 | Stardance
|
1793102041.2796018
|
1793706841.2796018
|
|
50454
|
idk by @YashaswaSrivastava | Stardance
|
YashaswaSrivastava
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @YashaswaSrivastava | Stardance
|
1793339694.263712
|
1793944494.263712
|
|
43585
|
idk by @Zakaria | Stardance
|
Zakaria
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @Zakaria | Stardance
|
1793228147.6219182
|
1793832947.6219182
|
|
62315
|
idk by @akhilatyam | Stardance
|
akhilatyam
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @akhilatyam | Stardance
|
1793652997.9736104
|
1794257797.9736104
|
|
18917
|
idk by @aqualz | Stardance
|
aqualz
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @aqualz | Stardance
|
1793375292.9803836
|
1793980092.9803836
|
|
44297
|
idk by @azimzihad | Stardance
|
azimzihad
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @azimzihad | Stardance
|
1793267258.2628143
|
1793872058.2628143
|
|
44261
|
idk by @chahatesh | Stardance
|
chahatesh
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @chahatesh | Stardance
|
1793454546.3093278
|
1794059346.3093278
|
|
50631
|
idk by @dnv-2012 | Stardance
|
dnv-2012
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @dnv-2012 | Stardance
|
1793353993.1282146
|
1793958793.1282146
|
|
72476
|
idk by @eshunburge | Stardance
|
eshunburge
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @eshunburge | Stardance
|
1793476043.0110776
|
1794080843.0110776
|
|
55417
|
idk by @hien | Stardance
|
hien
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @hien | Stardance
|
1793162939.7335985
|
1793767739.7335985
|
|
45019
|
idk by @il_giova2312 | Stardance
|
il_giova2312
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @il_giova2312 | Stardance
|
1793310565.2021623
|
1793915365.2021623
|
|
62195
|
idk by @imesh | Stardance
|
imesh
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @imesh | Stardance
|
1793652897.9255824
|
1794257697.9255824
|
|
29401
|
idk by @leevi | Stardance
|
leevi
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @leevi | Stardance
|
1793551907.582312
|
1794156707.582312
|
|
3616
|
idk by @lucia | Stardance
|
lucia
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @lucia | Stardance
|
1793231469.5703456
|
1793836269.5703456
|
|
71081
|
idk by @mnavarro30 | Stardance
|
mnavarro30
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @mnavarro30 | Stardance
|
1793530569.1843288
|
1794135369.1843288
|
|
49935
|
idk by @msaditi | Stardance
|
msaditi
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @msaditi | Stardance
|
1793314098.7984352
|
1793918898.7984352
|
|
57749
|
idk by @nai123 | Stardance
|
nai123
|
0
|
4.19
|
0
|
[]
|
[{"post_id": 63180, "ts": "202 [{"post_id": 63180, "ts": "2026-09-06T20:52:28Z", "hours": 0.2786, "text": "Restarting my Hackpad build again on SelfCAD… the modes kept switching and confusing me, and I kept adding blocks until I literally forgot how many I placed I’m not even sure if I’m going in the right direction, but after messing around with it for a bit I finally think I got the hang of it. Round 2, let’s see how far I get"}, {"post_id": 63024, "ts": "2026-09-06T16:17:52Z", "hours": 0.3428, "text": "Restarting my Hackpad AGAIN… now on SelfCAD. I tried making the 3×3 switch holes but SelfCAD started lagging, glitching, and refusing to let me duplicate cubes 💀 I had random cubes floating everywhere and the app was moving like it was underwater, so I cut the recording short before I lost my mind. Does anyone else’s SelfCAD act like this or is it just beefing me personally??"}, {"post_id": 62103, "ts": "2026-09-05T10:57:37Z", "hours": 0.5089, "text": "Worked on my Hackpad case again and honestly I don’t know what I’m doing 😭. I’m staring at Tinkercad wondering if I should just switch to something else because this app is beefing with me for no reason. I’m gonna start using the actual Hack Club guide because maybe that will stop me from losing braincells, and I’m trying to stop relying on AI every 2 seconds… but also I still don’t know if I need help or not. I’m just clicking things, aligning things, deleting things, praying things. The case is happening though. Slowly. Chaotically. I’m surviving."}, {"post_id": 61848, "ts": "2026-09-04T23:41:27Z", "hours": 0.3967, "text": "Restarted the case. Added cylinders. I’m just clicking things at this point and praying it becomes a Hackpad."}, {"post_id": 61431, "ts": "2026-09-04T10:43:42Z", "hours": 0.7275, "text": "Spent this whole session trying to line up the PCB area and I swear Tinkercad was fighting me the entire time. The view kept snapping weirdly, the layout kept shifting, and I had to redo the same thing like five times. I didn’t hollow the case yet because I was too busy trying not to throw my laptop, but at least I finally figured out where the PCB space needs to go. Next step: hollowing the case… if I survive."}, {"post_id": 60983, "ts": "2026-09-03T17:44:14Z", "hours": 0.9306, "text": "Okay so… I wasn’t supposed to be doing Part 2 yet but my brain said “nope, we’re building again.” It’s literally been like 2 hours since I posted Part 1, but I got bored and my inner self just wouldn’t let me stop. So here I am, already working on the case walls and base plate.I swear this part nearly finished me off. The ground kept moving away every time I tried to place something. I’d line it up perfectly, then Tinkercad would just slide the whole thing like it was on ice. I was actually getting frustrated because nothing wanted to stay still.But eventually I managed to get the base plate under the keys and started building the frame around it. It took way longer than it should’ve because everything kept shifting, but I got it done in the end.Anyway, here’s the progress. I’ll probably keep going because my brain refuses to chill. Good luck to me, honestly."}, {"post_id": 60912, "ts": "2026-09-03T15:55:34Z", "hours": 1.0006, "text": "Today I worked on my Hackpad case in Tinkercad, and honestly it was a whole mess. I kept messing things up, stopping the recording, Googling every two minutes, then coming back just to mess up again. It felt like I was stuck in a loop of “wait… what am I even doing.”The text on the keycaps nearly finished me. I was clicking everything except the right thing, scrolling in places that didn’t even scroll, opening random menus like I knew what I was doing. But eventually I figured it out, and it actually felt good when it finally worked. Anyway, I placed the keycaps and added text to some of them. I’ll finish the rest later after I mentally reboot myself. To whoever’s reading this… yeah, that’s all. Good luck to both of us lol."}]...
|
Title: idk by @nai123 | Stardance
Devlogs:
Today Title: idk by @nai123 | Stardance
Devlogs:
Today I worked on my Hackpad case in Tinkercad, and honestly it was a whole mess. I kept messing things up, stopping the recording, Googling every two minutes, then coming back just to mess up again. It felt like I was stuck in a loop of “wait… what am I even doing.”The text on the keycaps nearly finished me. I was clicking everything except the right thing, scrolling in places that didn’t even scroll, opening random menus like I knew what I was doing. But eventually I figured it out, and it actually felt good when it finally worked. Anyway, I placed the keycaps and added text to some of them. I’ll finish the rest later after I mentally reboot myself. To whoever’s reading this… yeah, that’s all. Good luck to both of us lol.
Okay so… I wasn’t supposed to be doing Part 2 yet but my brain said “nope, we’re building again.” It’s literally been like 2 hours since I posted Part 1, but I got bored and my inner self just wouldn’t let me stop. So here I am, already working on the case walls and base plate.I swear this part nearly finished me off. The ground kept moving away every time I tried to place something. I’d line it up perfectly, then Tinkercad would just slide the whole thing like it was on ice. I was actually getting frustrated because nothing wanted to stay still.But eventually I managed to get the base plate under the keys and started building the frame around it. It took way longer than it should’ve because everything kept shifting, but I got it done in the end.Anyway, here’s the progress. I’ll probably keep going because my brain refuses to chill. Good luck to me, honestly.
Spent this whole session trying to line up the PCB area and I swear Tinkercad was fighting me the entire time. The view kept snapping weirdly, the layout kept shifting, and I had to redo the same thing like five times. I didn’t hollow the case yet because I was too busy trying not to throw my laptop, but at least I finally figured out where the PCB space needs to go. Next step: hollowing the case… if I survive.
Restarted the case. Added cylinders. I’m just clicking things at this point and praying it becomes a Hackpad.
Worked on my Hackpad case again and honestly I don’t know what I’m doing 😭. I’m staring at Tinkercad wondering if I should just switch to something else because this app is beefing with me for no reason. I’m gonna start using the actual Hack Club guide because maybe that will stop me from losing braincells, and I’m trying to stop relying on AI every 2 seconds… but also I still don’t know if I need help or not. I’m just clicking things, aligning things, deleting things, praying things. The case is happening though. Slowly. Chaotically. I’m surviving.
Restarting my Hackpad AGAIN… now on SelfCAD. I tried making the 3×3 switch holes but SelfCAD started lagging, glitching, and refusing to let me duplicate cubes 💀 I had random cubes floating everywhere and the app was moving like it was underwater, so I cut the recording short before I lost my mind. Does anyone else’s SelfCAD act like this or is it just beefing me personally??
Restarting my Hackpad build again on SelfCAD… the modes kept switching and confusing me, and I kept adding blocks until I literally forgot how many I placed I’m not even sure if I’m going in the right direction, but after messing around with it for a bit I finally think I got the hang of it. Round 2, let’s see how far I get...
|
1793451330.91928
|
1794056130.91928
|
|
72558
|
idk by @nathanchandra | Stardance
|
nathanchandra
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @nathanchandra | Stardance
|
1793515628.8682985
|
1794120428.8682985
|
|
36217
|
idk by @renjiro | Stardance
|
renjiro
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @renjiro | Stardance
|
1793249569.1942997
|
1793854369.1942997
|
|
49892
|
idk by @rudrakshsharma | Stardance
|
rudrakshsharma
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @rudrakshsharma | Stardance
|
1793310592.2049804
|
1793915392.2049804
|
|
49893
|
idk by @rudrakshsharma | Stardance
|
rudrakshsharma
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @rudrakshsharma | Stardance
|
1793310593.1261582
|
1793915393.1261582
|
|
59424
|
idk by @samueldibu1 | Stardance
|
samueldibu1
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @samueldibu1 | Stardance
|
1793559089.9076915
|
1794163889.9076915
|
|
34170
|
idk by @snpdonutahhahah | Stardance
|
snpdonutahhahah
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @snpdonutahhahah | Stardance
|
1793195263.712828
|
1793800063.712828
|
|
3945
|
idk by @theLegend | Stardance
|
theLegend
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @theLegend | Stardance
|
1793494088.7129066
|
1794098888.7129066
|
|
4818
|
idk by @the_ryaan_ahmed | Stardance
|
the_ryaan_ahmed
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @the_ryaan_ahmed | Stardance
|
1793191898.6781664
|
1793796698.6781664
|
|
60459
|
idk by @turtle-key | Stardance
|
turtle-key
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk by @turtle-key | Stardance
|
1793641999.880448
|
1794246799.880448
|
|
71750
|
idk what im doing by @zaidabajwa | Stardance
|
zaidabajwa
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk what im doing by @zaidabajwa | Stardanc Title: idk what im doing by @zaidabajwa | Stardance...
|
1793191636.0222435
|
1793796436.0222435
|
|
62650
|
idk yet by @cellowill2318 | Stardance
|
cellowill2318
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: idk yet by @cellowill2318 | Stardance
|
1793656362.6488848
|
1794261162.6488848
|