Jovany Wang

Ant Group · AI Infra

I’m currently working in the area of optimizing AI infra performance at Ant Group’s AI Infra team. These years, I’ve been focusing on optimizing online inference and serving performance for search and recommendations, applying AI compilation techniques and other modern optimization tools. Some of the work is also about making Ant AI Infra more capable and more stable.

For the past 3–4 years, I worked on Ray as a Ray Core committer at Ant Group. I also led the Ray AI ecosystem there, including RayFed, RLlib, and Ray AI Runtime. Ray is a high-performance distributed AI computing system.

I’m a follower of modern C++, and I believe in zero-cost abstraction. My email is kingchin1218@gmail.com — feel free to write to me if you have any issue.

Interests

  • AI compilation
  • LLM infra
  • Performance optimization