Sean Moran

I am an AI researcher and engineer based in London. I work on information retrieval, code intelligence, reliable AI systems, and machine learning for settings where privacy, auditability, and operational constraints matter.

I previously led an applied AI research group at JPMorgan Chase. Before that, I was a senior research scientist at Huawei’s research centre in London. I received my PhD from the University of Edinburgh, where I was a member of EdinburghNLP.

My work has resulted in more than 40 peer-reviewed papers and 25 granted US patents. The complete records are on Google Scholar and Google Patents.

Selected research

  1. CodeQUEST: Evaluation and Improvement of Code Quality Using LLMs

    R. Liu, A. Frade, A. Vaidya, M. Labonne, M. Kaiser, B. Chakrabarti, S. Moran. ISSREW, 2025.

    code

  2. DeepClean: Machine Unlearning by Resetting Privacy-Sensitive Weights

    J. Shi, N. Ghalyan, K. Gourgoulias, J. Buford, S. Moran. ECCV Workshop on Unlearning, 2024.

  3. Spam-T5: Benchmarking LLMs for Few-Shot Email Spam Detection

    M. Labonne, S. Moran. FinLLM at IJCAI, 2023.

    code

  4. Senatus: A Fast and Accurate Code-to-Code Recommendation Engine

    F. Silavong, S. Moran, A. Georgiadis, R. Saphal, R. Otter. MSR, 2022.

  5. DeepLPF: Deep Local Parametric Filters for Image Enhancement

    S. Moran, P. Marza, S. McDonagh, S. Parisot, G. Slabaugh. CVPR, 2020.

    codevideo

  6. Low-Light Video Enhancement Using Synthetic Data with Intermediate Domain Mapping

    D. Triantafyllidou, S. Moran, S. McDonagh, S. Parisot, G. Slabaugh. ECCV, 2020.

    codevideo

  7. Sparse Kernel Learning for Image Annotation

    S. Moran, V. Lavrenko. ICMR, 2014. Best Student Paper.

    code

  8. Variable Bit Quantisation for LSH

    S. Moran, V. Lavrenko, M. Osborne. ACL, 2013.

    code

All publications

Systems and code

  • BitBudget — a benchmark for measuring embedding retrieval quality per byte. Code.
  • The Corpus is the Model — a training-free image tagger whose predictions can be traced to examples in the corpus.
  • API-Miner — probabilistic retrieval of related API specifications.
  • CV4Code — source-code understanding through spatial representations.

Talks, code, and datasets

Recent writing

All writing