About

Professional Bio

I am a researcher and engineer currently in Google DeepMind, specializing in NLP and a contributor to the Gemini series of models. My work focuses on the intersection of reinforcement learning algorithms and infrastructure. Now I am also interested in robotics.

🎓 Selected Publications

For a full list of my work, please visit my Google Scholar Profile.

• Q-BERT: Hessian Based Ultra Low Precision Quantization of BERT
  NeurIPS, 2019

• Boosting Agentic Reasoning in LLM Judges via Tool-Integrated Reinforcement Learning
  arXiv, 2024

• AutoHoot: Automatic High-Order Optimization for Tensors
  MLSys, 2020

• Inefficiency of K-FAC for Large Batch Size Training
  arXiv, 2019

• Gemini 2.5: Pushing the frontier with advanced reasoning, multimodality, long context, and next generation agentic capabilities
  Google DeepMind, 2024

• Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context
  Google DeepMind, 2024

• Gemini: A Family of Highly Capable Multimodal Models
  Google DeepMind, 2023