Tianle Li

head_shot.jpg

firstlast@berkeley.edu

I am a Member of Technical Staff at xAI, working on reasoning, post-training, and RL.

  • Core contributor to Grok 4
  • Co-creator of Grok 4 Mini: lead post-training RL training and recipes, co-lead distillation.

Previously, I was an EECS undergraduate at UC Berkeley, where I was fortunated to be advised by Ion Stoica and built LMArena. During my undergrad, I also spent a year full-time at Nexusflow as part of the LLM post-training team, collaborating with Banghua Zhu and Jiantao Jiao. Additionally, I briefly worked as a student researcher at Google AI Research, on reasoning.

I am very interested in fundamental problems in training large models, building more capable and reliable models, and solving superintelligence. Some of the concrete problems I have been thinking about recently:

  1. How can we transfer intelligence learned in verifiable domains to open-ended questions?
  2. Design and formulate less hackable and more interpretable reward signals for presentation and style.
  3. How can we leverage multi-agents to train smarter models.