GPU fleet monitoring for local inference. Tracks tok/W, WES score, thermal state, and routing decisions across multi-node setups. The observability layer that makes local-first inference trustworthy. Community tier free.
Personal AI OS. Sovereign, local-first, runs on your hardware. Knows your goals, monitors your world, scales across your whole life — chief of staff, life coach, intel analyst, publisher. Built on compass-md. Coming soon.
Open specification for portable AI context — the files that tell any AI tool who you are, how you work, and what you care about. MIT licensed. Works with Claude Code, Cursor, local models, anything that reads files.
Local dashboard for reviewing compass-md refinements. Accept, reject, and apply agent-proposed context changes without leaving your machine. Go backend, runs at localhost:7174.
Open methodology for building private enterprise language models. The thesis:
a 7B model trained on the right data beats a frontier model on a narrow task
and runs locally for free. Current work: account-intelligence-7b-v1 —
fine-tuned Qwen2.5-7B for enterprise account briefs across six surfaces.
Agentic enterprise platform running Qwen2.5-32B-FP8 on a DGX Spark alongside Claude. Generates account intelligence, meeting prep, and pricing assistance for sales teams. Peers into Salesforce, Confluence, and internal systems in real time.