Tianqi Chen's team open-sources a book on compiler-driven agentic GPU kernel optimization for MLSys
sh_reya · x · 2026-09-30
Tianqi Chen's team will teach "Agentic GPU Programming for MLSys" in CMU's ML Systems course this spring and released a companion open-source online book.
- Core idea: coding agents increasingly handle GPU kernel work — finding implementations, writing kernels, interpreting diagnostics, and choosing what to try next.
- Harness: a compiler-driven environment with domain-specific program analysis, a knowledge base, and GPU benchmarking/profiling, letting agents express optimization ideas, diagnose errors, measure performance, and retain worthwhile results.
- Structure: Part I covers the compiler harness and how agent workflows compose with it, touching on knowledge curation, compiler analysis, and feedback loops.
- It is a companion to Modern GPU Programming for MLSys and will be integrated into the CMU MLSys course series; contributions are welcome on GitHub.
More from coding & agent
- Codex-built Mario-style game compiles to .gbc and runs on a real Game Boy Color — pvncher · 2026-09-30
- Free hands-on guide: turning AI coding into a full-lifecycle 'Agentic Software Factory' — Pavan_Belagatti · 2026-09-30
- Hugging Face Details Moon Bot, Its Slack-Native Coding Agent Built on HF Buckets — lhoestq · 2026-09-30
- Hugging Face Engineer Shares Full Slides on How the Company Uses Agents Internally — lhoestq · 2026-09-30
- DynaRobotics' full-stack approach to workflow automation wins industry praise — chris_j_paxton · 2026-09-30
- STILL, a game that's 100% vibe coded, shows fully interactive city gameplay — Daniel_Farinax · 2026-09-30