frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Open in hackernews

Show HN: Watch 14-Byte AI "brains" attempt to solve a 2D maze (Its hard)

https://con-dog.github.io/MINIMIO-PUBLIC-FRONTEND/
3•purple-leafy•1h ago
Hey HackerNews,

I built this project over the last few weeks as a palette cleanser from a failed game launch.

I wanted to learn a bit about AI/Neural-Networks and naively thought I could build a tiny maze-solving AI in a weekend with a 100% solve rate.

Well - I couldn't, but I got pretty close. 14 Bytes total model size, and a 96.5% solve rate on unseen mazes. Trained across 46 phases experimenting with different ideas to improve the model (better performance, smaller size).

Its quite fun to watch the model attempt to solve the maze, when they fail its usually due to getting stuck in a loop. The models have no access to coordinates, map-data, or external memory scratches - they must navigate using only immediate local neighbourhood observations.

There is a model dropdown and you can see how the model has progressed over each phase, constantly getting smaller and increasing its solve rate. Total trained models number in the thousands - I just expose the winning models from each phase.

Overall a fun experiment, with much implementation help from AI agents to scaffold and implement the code (I'm a lazy software dev).

Comments

purple-leafy•1h ago
Facts:

- 14 Bytes does not include the runtime! - All that counts towards the 14 Bytes is the model weights - Lack of git history is because my actual repo is private, this is just the frontend for now + models

Why does everything feel so joyless?

https://www.theguardian.com/us-news/ng-interactive/2026/jul/26/age-of-decadence-pleasure-ai
1•mattjhall•1m ago•0 comments

Ask HN: Where do those LLM-generated outreach emails come from?

2•nicbou•4m ago•0 comments

Modern email can be built from borrowed parts

https://en.andros.dev/blog/d7ed8b07/modern-email-can-be-built-from-borrowed-parts/
1•andros•7m ago•0 comments

Would you choose £50k over the chance of £1M?

https://yougov.com/en-gb/daily-results/20260722-c0b92-3
2•yzydserd•10m ago•1 comments

Elevated errors on Claude Opus 5

https://status.claude.com/incidents/lhqp09kxq7pb
2•flyaway123•14m ago•0 comments

Sam Altman says we are in the singularity

https://www.businessinsider.com/sam-altman-openai-the-singularity-agi-prediction-anthropic-nvidia...
1•ColinEberhardt•15m ago•1 comments

Show HN: adCasa OS – AI marketing workspace built with Bayesian attribution

https://adcasa.io/
1•adcasa•16m ago•0 comments

Nvidia in talks with OpenAI to guarantee $250B financing for data center

https://www.reuters.com/business/media-telecom/nvidia-talks-with-openai-guarantee-250-billion-fin...
1•01-_-•17m ago•0 comments

Zeiss expands German site that caps ASML's EUV scanner output

https://www.tomshardware.com/tech-industry/zeiss-expands-german-site-that-caps-asmls-euv-scanner-...
1•01-_-•19m ago•0 comments

Doom Running on a Regex Engine

https://4rh1t3ct0r7.github.io/doom-regex/
1•abareplace•21m ago•0 comments

Meta launched a new AI optimism ad set to a song about human extinction

https://techcrunch.com/2026/07/23/meta-launched-a-new-ai-optimism-ad-set-to-a-song-about-human-ex...
4•robin_reala•25m ago•0 comments

Do Agent Benchmarks Measure Capability? Protocol Validity in the Age of Agentic

https://arxiv.org/abs/2607.22368
1•sbulaev•27m ago•0 comments

I'm running the ICFP programming contest

https://eieio.games/blog/im-running-the-icfp-programming-contest/
1•birdculture•28m ago•0 comments

The AI bioweapon risk isn't jailbreaks

https://point.free/blog/the-ai-bioweapon-risk-isnt-jailbreaks/
1•cafkafk•29m ago•1 comments

Keigo Higashino, author of The Devotion of Suspect X, dies at 68

https://www.oricon.co.jp/news/2470322/
2•linzhangrun•37m ago•0 comments

Even in Utah Families Are Getting Smaller

https://www.nytimes.com/2026/07/25/us/politics/utah-birth-rate-fertility-decline.html
1•geox•45m ago•0 comments

Show HN: A browser-based vector animation engine rendering animated objects

https://thorvg-janitor.vercel.app/
3•hermet•47m ago•2 comments

Show HN: I made a convenient and fast file sharing service

https://coldfiles.dev/
2•bunchy7s•47m ago•5 comments

Show HN: Comparing stock returns after cyber incidents

https://incidentimpact.com/
2•0xad•52m ago•0 comments

Build internal tools for your Startup with Construct AI

https://construct.computer/blog/build-internal-tools-with-construct/
1•ankushKun•53m ago•0 comments

From Seed to Startup: How AI Is Revolutionizing Cannabis Cultivation

https://www.researchgate.net/publication/410602482_From_Seed_to_Startup_How_AI_Is_Revolutionizing...
1•fofoz•1h ago•0 comments

Could a Neuroscientist Understand a Microprocessor?(2017)

http://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1005268
1•num42•1h ago•0 comments

Can a Biologist Fix a Radio?(2002)

https://www.cell.com/action/showPdf?pii=S1535-6108%2802%2900133-2
1•num42•1h ago•0 comments

Reverse Engineering of Biological Complexity(2002) [pdf]

https://courses.physics.illinois.edu/phys498BIO/sp2009/LECTURE7/Doyle.pdf
1•num42•1h ago•0 comments

Ask HN: Does the tech industry require deterministic ML training?

3•Swapnopam•1h ago•0 comments

The Lean Theorem Prover: Design, Evolution, and Impact

https://leodemoura.github.io/static/floc26/
1•tosh•1h ago•0 comments

General Resolution: LLM Usage in Debian

https://www.debian.org/vote/2026/vote_002#texta
2•olalonde•1h ago•0 comments

LVMH CEO Bernard Arnault's open letter responding to Le Monde's investigation

https://twitter.com/i/status/2081439415268790328
1•Michelangelo11•1h ago•0 comments

Building Real-Time Financial Analytics with Stream Processing

https://medium.com/@DolphinDB_Inc/why-stream-processing-matters-and-how-dolphindb-does-it-b63c5dc...
2•chaercling•1h ago•0 comments

CstWeave – On-device, privacy-first Chrome extension for tab region GIFs

1•fujidevv•1h ago•0 comments