James Collier

Python Software Engineer with 10+ years of professional experience spanning software development, data science, machine learning and generative AI. Experienced in developing Python applications and APIs, integrating data systems and building automated workflows. Combines practical engineering experience with strong analytical and computational skills, underpinned by a PhD in Plasma Physics.

SOFTWARE ENGINEER (CONTRACT) - Percept - Remote

(2026-06)

• Contributed to a Python-based URL health checker for client advertising URLs, implementing crawler behaviour in accordance with RFC 9309 and robots.txt directives.
• Integrated an additional provider into an established Python advertising data pipeline framework, automating daily report retrieval and ingestion of campaign hierarchies into the data warehouse, and implementing the associated campaign-health metrics and insights.
• Evaluated approaches to predicting advertising campaign efficiency gains and detecting anomalies across campaign metrics, comparing methods and assessing their suitability.

SOFTWARE ENGINEER - EntityX - Remote

(2025-10 - 2026-06)

• Maintained and enhanced Streamlit applications, developing a new page for scheduled report configuration and optimising SQL queries against BigQuery.
• Developed FastAPI endpoints to serve scheduled report outputs to a React frontend.

SENIOR DATA SCIENTIST - Allianz - Guildford, Surrey

(2025-01 - 2025-10)

DATA SCIENTIST - Allianz - Guildford, Surrey

(2018-11 - 2025-01)

• Developed & deployed a GPT-4o-based email classification model, achieving 80% accuracy on a held-out test set using 1,000 business-labelled emails.
• Contributed to the team’s first RAG solution, developing Python-based document chunking methods to preserve section hierarchy and heading context. Later led development of a successor chatbot for underwriters, extending document ingestion to Word documents and replacing the file-based FAISS index with pgvector.
• Led development and deployment of two Gradient Boosting models, designing one of the models and a single API endpoint to expose both to a third-party system while allowing for future model additions.
• Managed the release of an email analytics and integration project, contributing code and resolving architectural challenges in a restricted, regulated environment.

DATA SPECIALIST - United Vanning - Staines-upon-Thames, Surrey

(2016-05 - 2018-09)

• Delivered consulting projects spanning Data Vault 2.0 warehousing, Master Data Management, and automated pipelines integrating multiple business systems.

PHD - PLASMA PHYSICS - University of Warwick

MASTER OF PHYSICS - University of Southampton