Research meets implementation
AI systems. Agent engineering.
Open-source computing.
I work on efficient model inference, agent systems and numerical computing. I lead the Owl open-source project and build systems that connect research ideas with practical applications.
PhD in Computer Science, University of Cambridge.
Research experience at KIT and Beijing Institute of Technology.
A scientific computing and deep learning framework for OCaml. My work spans numerical operators, computation graphs, memory optimisation, model interoperability and community maintenance.
86.6% lower Mask R-CNN CPU inference memory with graph-level memory reuse, compared with the same framework without graph optimisation.
Explore OwlA research project on timely inference for driving models. I designed a common model-adaptation and replay-evaluation framework and used controlled experiments to investigate performance.
17.9% lower median per-frame replay service time in an offline development benchmark, with output-consistency checks.
Explore OverDriveA legal knowledge system combining hybrid retrieval, entity knowledge and MCP tools. Designed, developed and deployed on AWS, with feedback from a pilot at a leading Chinese law firm.
Incremental knowledge maintenance, source-checked quotations and asynchronous processing; server costs reduced by approximately 67%.
Explore KnowledgeBase
Parallel execution and memory efficiency for learning under edge-device resource constraints.
A probabilistic synchronisation approach for heterogeneous edge learning.
Coordinating individual policies and collective behaviour in air-ground sensing tasks.
Beyond the code
I enjoy explaining systems as much as building them: through open-source documentation, books, teaching and collaboration. Away from the screen, I practise kendo, cycle and explore new places.
More about me →