Mehdi Jafarnia

I am a Research Engineer at Google DeepMind in Mountain View, CA. My work centers on reinforcement learning, post-training, and foundation models. I investigate novel uncertainty estimation and exploration paradigms for Large Language Models (LLMs), designing methods that improve model reasoning, data efficiency, and reliability in complex environments.

Prior to Google DeepMind, I spent a year at Uber Technologies, where I worked on dynamic pricing and cost estimation models.

I received my Ph.D. in Electrical Engineering from the University of Southern California (USC), alongside concurrent Master's degrees in Computer Science and Applied Mathematics. Prior to USC, I completed dual B.S. degrees in Electrical Engineering and Computer Science at Sharif University of Technology.