MComp Computer Science @ University of East Anglia. Software Engineering Intern @ ProdAxon, building production AI extraction pipelines on AWS.
Stack: Python · TypeScript · Django · React · PostgreSQL · AWS · Docker
Building a manufacturing intelligence platform that turns factory root cause analysis documents into structured, searchable knowledge.
- Document-AI extraction pipeline: OCR with confidence-based model routing, LLM segmentation and causal-graph extraction, deployed serverlessly on AWS Lambda
- Entity resolution over free-text engineer reports using pgvector embedding search, advisory-lock concurrency safety, and LLM adjudication for ambiguous matches
- Cross-language evidence matching linking maintenance fixes to the equipment-manual pages that support them
- Full-stack: React and TypeScript graph interfaces over Django and PostgreSQL
Code is in private repositories.
search-engine Four-stage information retrieval pipeline: scraping, preprocessing, inverted index, and TF-IDF ranking with NER query expansion and Precision@10 evaluation.
pathfinding-visualiser Interactive BFS, DFS and A* visualiser with a benchmark mode comparing runtime and nodes explored across algorithms.
Also a national-level sprinter and Track Captain at UEA Athletics.