4 projects
nvinx
nginx-style workload scheduler for limited-VRAM GPU bench setups
bentoloop
Parallel LLM scout + confidence-gated synthesis loop for GitHub-grounded research
bentocall
Recursive long-context LLM calls in lunchbox shape — cheaper than one big call when the model would drift on long context.
spa-ai
SPA AI — install Jidoka halts as PRs. A broken thread must stop the loom.
