Shilin He

Coding agents can achieve superior repository exploration, outperforming classical methods by effectively leveraging line-level context for bug diagnosis and code retrieval.

Proprietary MLLMs generate visually appealing code from Figma designs, but fall short on crucial aspects like responsive layouts and maintainability, revealing a surprising gap in their practical utility.

Automating software repository build and testing across languages and platforms is now possible, unlocking scalable benchmarking and training for coding agents.