A tool like Claude Code, Codex, etc. but designed from the beginning to work well with open and local models:

https://swival.dev

It also comes with nice features and benchmarking abilities. For running evals, it has a companion called Calibra https://calibra.swival.dev

Been recently playing around with using LLMs and the promise of malleable software.

Published a demo/experiment under MalleableTodo [1] - and so far seen some pretty strange use cases...

Essentially, just allows each user to use an LLM to rewrite their own UI to add features/customisation.

[1] https://malleabletodo.app/

Working on Functional Universe (FU), a formal framework for modeling physical reality as functional state evolution, integrating sequential composition with simultaneous aggregation.

https://voxleone.github.io/FunctionalUniverse/

Working on a Vercel-like cloud hosting PaaS, but specifically tailored to South Africa. Everything is hosted on local servers, and pricing is ZAR instead of USD. It's called Zanode if you want to check it out :)

https://www.zanode.co.za/

Started on making my own AI model benchmarks and leaderboard[0], after I tested MiniMax M2.5, which was supposedly good based on standard benchmarks, but peformed really poorly in practice and burned through hundreds of thousands of reasoning tokens for each request...

[0]: https://aibenchy.com

Building a boring POS (1) using various AI tools just to check what can I do with these tools. I have used claude, gemini and now using antigravity. I have not done a single edit manually.

I got it all done in probably an hour or two. But done in 10-15 min blocks over many days.

(1) https://pos.unlimit.in

I've been working on Peachy, a framework for writing native Linux applications easier and faster with React and GTK: https://peachy-9b8f81.pages.gitlab.gnome.org/

Very WIP and no docs, but I hope it will be helpful someday

Current project is https://mattera.law which is an inbox-centric matter/case management system for law firms.

Under the hood it uses a cool legal reasoning agent primarily designed for understanding litigation claims and objectives.

Sometimes I do wish I had a slack channel of like 30 attorneys so I can ask them questions and get feedback.

An iOS app size analysis tool you can run on your Mac and track build size changes over time https://apps.apple.com/us/app/dotipa/id6742254881

A fully vibe coded python 3.14 interpreter in Rust: https://blueblazin.github.io/pyrs

Now at 350k lines. Native and wasm binaries (you can try the limited wasm version online). Currently adding a full CPython test suite benchmark.

Just for fun, not trying to replace CPython here. Mainly to test the limits of current coding agents.