Building local-first AI systems, agent frameworks, and experimental tooling.
Projects ready for use.
- pi-qwen-coding-plan - pi mono and oh-my-pi qwen coding plan endpoint config (original work)
- event-invariant-language - spec for agent handoff invariants (original work)
- hard-constraints-acceptance-criteria - spec for agents to complete work (original work)
- lmstudio_hf - creates symlnks from Hugging Face cache to LM Studio folder (fork)
Currently evolving projects I'm adapting to my use.
- oh-my-pi (fork)
- oh-my-pi-crust - rust port (originalish work)
- beads (fork)
- persistent-mind-model-v1.0 (fork)
Things I'd like to work on.
- mlx-lm (fork)
- mlx-vlm (fork)
- continuous-thought-machines (fork)
- icestudio (fork)
- BDHs (fork)
- bio_inspired_nanochat (fork)
Things I need to study.
- pi_agent_rust (fork)
- pi-mono (fork)
- ypi (fork)
- oh-my-opencode (fork)
- mergekit (fork)
- recursive-llm (fork)
- rlm (fork)
- rlm-minimal (fork)
- MX-QLLM (fork)
- HelpBookDemo (fork)
- dgm (fork)
- raplh-playbook
- lambda-lang
- serax