Can we use machine learning to create better tools for software engineers? My research concerns machine learning models and methods that "understand" and generate source code. By learning from existing code, I aim to create useful machine learning-based software engineering tools, interfaces and insights. My research focuses on developer tools with a strong machine learning component, while using problems in this area to motivate machine learning research.
I am currently a researcher at Microsoft Research in Cambridge, UK and part of the Deep Program Understanding project.