I work to make software development easier and more accessible so that every programmer can easily write, maintain and optimise software. I maintain a list of must-read research papers for machine learning in compilers. I am interested in solving real-world problems by building working prototypes to be tested in real-life environments and real computing hardware using realistic workloads. We are a strong believer in reproducibility with a track record of engaging with the artefact evaluation process of major systems conferences. My research interests include:

• Program optimisation and analysis: I am interested in how we can design new algorithms and optimisation techniques at compilers and operating systems to improve application performance and energy efficiency.
• Software testing and software reliability: I use machine learning and code analysis techniques to detect software bugs and to improve software reliability; Our recent work has identified 200+ bugs from real-life projects.
• Accelerate large-scale deep learning models: I work to make the training and use of large-scale deep learning models accessible to every data scientist by collaborating with major industrial players including Meta.
• Applied machine learning: Some of my works apply machine learning to emerging applications areas like natural language processing, data mining and wireless sensing.

I am a (full) professor and a member of the Distributed Systems and Services Group at Leeds, the EPSRC Peer Review College, the UKRI Talent Peer Review College (PRC), and the HiPEAC Network of Excellence. I was a Turing Fellow at The Alan Turing Institute. I was named among the Elsevier and Stanford World’s Top 2% Scientists in 2020, 2021, 2022, 2023 and 2024. I am a recipient of the Test-of-time Award in CGO 2024, Best Paper Award in ACM PACT 2010 and 2017, ACM CGO 2017 and 2019, Best Presentation Award at PACT 2010 and CGO 2013, 4 HiPEAC paper awards, and Best Paper Nomination/Finalist in ACM SC 2024, ACM SenSys 2019, and ACM CCS 2018. I am in the unofficial CGO hall of fame. I have an Erdős Number and a Dijkstra number of four.