Zhe Li

Member of Technical Staff, Research

Zhe works on model training and RL environment design at Patronus AI, where he develops the infrastructure and training recipes for post-training models using supervised fine-tuning (SFT) and RL. Prior to joining Patronus AI, he was a Member of Technical Staff at Inflection AI, focusing on large-scale model training, inference, and infrastructure. Previously, he was a Staff Machine Learning Engineer at Apple, where he worked on Face ID and Apple Vision Pro. He holds a PhD from the University of Iowa.

At Patronus· Newest first

Newest first

Papers

No items found.

Posts

Valley of Ten Thousand Smokes, Alaska. Photo: National Park Service (public domain)

We Built Our Own Post-Training Stack to Measure What Our Tasks Are Worth

Oct 29, 2026
·
Engineering
post-training-stack
Zhe Li
Evan Jolley
Bartosz Mielczarek
Abdelrahman Madkour
Supercell over Vrana Lake, Croatia. Photo: Šime Barešić (CC0)

Getting GLM-5.2 NVFP4 Post-Training Off the Ground

Jul 8, 2026
·
Engineering
getting-glm-5-2-nvfp4-post-training-off-the-ground
Varun Prashant Gangal
Zhe Li
Rain over the tallgrass prairie, Tallgrass Prairie National Preserve, Kansas

No papers or posts here yet.

Work with Zhe

See open roles →
Zhe Li
show