Andrew Docherty is a PhD-qualified researcher and software engineer based in Sydney with 16 years of experience applying fundamental mathematics and advanced computational methods across academic, corporate R&D, and startups. He specialises in graph machine learning, computer vision, and NLP (LLMs/Transformers), and has recently led self-directed research into LLM generation processes to improve AI reliability. Comfortable bridging theory and practice, he has contributed practical examples to the StellarGraph project—implementing a HinSAGE MovieLens recommender with data pipelines, Redis integration, evaluation metrics, and production-ready scripts. His work spans modelling complex physical systems (notably optical devices) to probing the internal mechanics of modern AI, demonstrating both deep technical rigor and an appetite for reproducible, applied research.
Contributions:3 releases, 4 reviews, 276 commits in 3 months
Contributions summary:Andrew implemented an HinSAGE example for the MovieLens recommender system, which involved writing helper functions to write MovieLens data to Redis and created the necessary schemas. The user also added example scripts. This includes creating graph data structures, processing, and integration of a model. The work included implementing evaluation metrics and creating a framework for running the model and saving predictions, as well as cleaning up code.
Automatically exported from code.google.com/p/polymode
Contributions:29 commits, 3 PRs, 3 pushes in 12 years 4 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.