Bio
Researcher at Applied Compute. Building the post-training stack, training specialized workhorse models for enterprises, and researching new techniques for model customization. Graduated from MIT.
Session (1)
Additional Speaker Details
Using publicly available information we constructed an analysis to help you get a feel for this speaker before deciding to attend their session.
Raymond Feng is a software engineer at Applied Compute, a startup building "specific intelligence" for enterprise AI agents through post-training, RL environments, and continual learning on proprietary production data. He co-authored Applied Compute's June 2026 collaboration with Harvey on post-training a frontier legal agent, so his session should offer a practitioner's view of the work at the boundary of evals, reinforcement learning, grader design, and production model improvement.