Flint runs auto-research on a laptop: 26% faster scheduling vs round-robin
cong_ml · x · 2026-10-22
congml demos Recursive Flint running a full automated research loop locally on a laptop, using only code and scores — no vision needed.
- Job-shop scheduling: Flint wrote new ways to reorder work and search for better schedules, finishing 26% earlier than round-robin on the same jobs and machines
- Circle packing (previous post): the model jointly moves and resizes circles, polishes layouts and tries new starts, with surprisingly good results
The author frames this local auto-research as bringing the magic of efforts like OpenAI's Navier–Stokes work and AlphaEvolve to everyone's own machine.
Related event: Recursive Flint runs autonomous research on a laptop, open-sources 40B-A2B(7 posts)→
More from coding & agent
- Migrating an entire Hermes automation workload to Antigravity took about 2 hours — rseroter · 2026-10-22
- How a second Mac mini lets AI agents auto-distribute compute-heavy tasks — tedddyoweh · 2026-10-22
- LangChain's Design Agent Shipped 3,000+ Assets and Earned 4.5M Views Since August — LangChain · 2026-10-22
- Databricks Chief AI Scientist: Evals Are the Biggest Bottleneck to Enterprise AI Adoption — jefrankle · 2026-10-22
- Developer plans a solar-aware job scheduler so his homelab rides free excess power — mertdumenci · 2026-10-22
- How do you design your agentic code review workflow? — lucasmeijer · 2026-10-22