AI Agents Optimize Scientific Code: Electronic DFT Gets 36x Speedup
rishabh16_ · x · 2026-11-13
A computational chemistry researcher discusses the trend of using AI agents to optimize scientific computing code. Agents have already been deployed to peruse codebases and fix bugs (e.g., a softmax/cross-entropy bug in torch/jax), and a recent 36x speedup in electronic DFT by @discoveredmat adds momentum. The author looks forward to agents optimally re-implementing forcefields/potentials for MD and FEP, expecting major speedups still waiting to be uncovered.
Related event: AI agents speed up scientific computing, boosting DFT 36x(3 posts)→
More from coding & agent
- SGLang Summit: agents now write ~95% of merged PRs as v1.0 and Miles v0.2 ship — hsu_byron · 2026-11-13
- Real-time chess tracking from iPhone video with RF-DETR and Stockfish — measure_plan · 2026-11-13
- OmniMind: an open-source MCP server for handing off tasks between AI agents — IrfanCodesBTW · 2026-11-13
- Claude Design Crit in Canvas: Screenshots Every Component State, Fixes All Comments in Parallel — max__drake · 2026-11-13
- Billion-dollar Unity ships the same editor for 10 years; one user builds their own equivalent — gdechichi · 2026-11-13
- When anyone can vibecode, the moat is UX polish, says ex-Google VC — vivekhaldar · 2026-11-13