localcode
pip-installable AI coding agent that runs entirely on Apple Silicon. no cloud, no API keys, no data leaving your machine.
Problem
Cloud-only AI coding assistants impose three taxes. Privacy — every prompt and code snippet leaves your machine. Latency — every keystroke round-trips to someone else's data centre. Cost — pay a subscription or get rate-limited. None of these are fundamental: Apple Silicon can run capable coding models locally if someone bothers to package it well.
Approach
A pip-installable AI coding agent that runs entirely on your Mac. First launch builds the inference server and downloads the model (~5 min, one-time). After that, startup is ~15 seconds.
- Native Apple Silicon — MLX runtime, no Docker, no Python venv hell beyond the pip install itself.
- Owns its own tools — reads + edits files (refuses destructive overwrites), runs commands (auto-detects long-running servers and backgrounds them), searches code, builds and launches apps.
- Adaptive thinking — uses reasoning for planning + debugging, skips it for routine codegen.
→ github.com/mjwsolo/localcode
Today
Open source, alpha. Active development — expect rough edges, breaking changes between versions, and bugs. Looking for early users who run on Apple Silicon and don't mind filing issues.