Modal: RL Training of Your Own Agent Models Is Now Practical
AI Engineer · youtube · 2026-11-09
Modal's Jason Mancuso argues treating models as swappable backends is a losing strategy—teams with agents in production already have what RL needs: a harness, evals and real traces. He explains why RL generalizes better than SFT on recorded traces, then walks through a stable async RL recipe built on GRPO with masked importance sampling, score centering and top-p sampling. Sparse weight updates let Modal spread rollouts across clouds, GPU types and quantization levels while training stays stable and rollouts get 15% cheaper.
More from coding & agent
- GitNexus Turns Any GitHub Repo Into an In-Browser Interactive Knowledge Graph, Fully Open-Source — JafarNajafov · 2026-11-09
- Better Paseo plugin puts every AI agent on one task board across phone, iPad and desktop — intellectronica · 2026-11-09
- Alibaba open-sources Zvec, an in-process vector DB that's already at 16.2k GitHub stars — bigaiguy · 2026-11-09
- Keep, the save-as-Markdown tool for AI agents, launches on iOS after 2-month App Store review — iannuttall · 2026-11-09
- Claude Code cross-session messaging in parallel worktrees: two limits and a hardening config — Independent-Sir3234 · 2026-11-09
- Open-source RAG eval tool ships recall, faithfulness and cost metrics with no API key — infrixo · 2026-11-09