Dan Zheng
AboutBlogResume

Me

Welcome!

I’m interested in programming languages and machine learning for code.

Currently, I work on machine learning for code and software engineering at Google DeepMind.

Previously, I worked on Swift for TensorFlow at Google Brain, focusing on differentiable programming in Swift.

Selected works

Google Scholar

  • Gemini 2.5: Pushing the frontier with advanced reasoning, multimodality, long context, and next generation agentic capabilities - arXiv 2025
  • From Naptime to Big Sleep: Using Large Language Models To Catch Vulnerabilities In Real-World Code - Project Zero blog 2024
  • AI-assisted Assessment of Coding Practices in Industrial Code Review - AIware 2024
  • Dynamic Inference of Likely Symbolic Tensor Shapes in Python Machine Learning Programs - ICSE 2024
  • Resolving Code Review Comments with Machine Learning - ICSE 2024
  • DIDACT: Large sequence models for software development activities - Google Research blog 2023
  • Static prediction of runtime errors by learning to execute programs with external resource descriptions - ICLR 2023
  • Implementation Strategies for Mutable Value Semantics - Journal of Object Technology 2022
  • Swift for TensorFlow: A portable, flexible platform for deep learning - MLSys 2021
  • Demystifying differentiable programming: Shift/reset the penultimate backpropagator - ICFP 2019
me@danzheng.me