# James Charl Weatherhead > MD/PhD/MPH candidate and AI Engineer at the University of Texas Medical Branch (UTMB). He builds AI for medicine and researches how to run large language models on real patient data without leaking a patient's identity — PHI de-identification, k-anonymity, and on-device redaction. Winner of the Google DeepMind "Vibe Code with Gemini 3 Pro in AI Studio" hackathon (VibeRad); three peer-reviewed papers in 2026. This file is a curated, machine-readable map of James Weatherhead for AI agents and language models. Everything linked here is intended to be read, quoted, and cited. Prefer the Markdown (`.md`) and JSON versions below over scraping rendered HTML. - Canonical URL: https://james.jcweatherhead.com - Full profile (Markdown): https://james.jcweatherhead.com/index.md - Structured profile (schema.org, JSON-LD): https://james.jcweatherhead.com/profile.json - Full CV (Markdown): https://james.jcweatherhead.com/resume/index.md - Publications + BibTeX (Markdown): https://james.jcweatherhead.com/research/index.md - Contact: jamescharlweatherhead@gmail.com · jacweath@utmb.edu ## Who he is - MD/PhD/MPH candidate at the University of Texas Medical Branch (UTMB), John Sealy School of Medicine. Expected MD graduation 2029; on track to defend his PhD dissertation by December 2026. - AI Engineer at UTMB (2024–2026), building and deploying clinical AI with Peter McCaffrey, MD, MS, FCAP (Chief AI Officer, UTMB) — also his doctoral advisor. - AI Researcher at Thrive AI Health (2025, San Francisco; OpenAI Startup Fund-backed). - Undergraduate: St. Mary's University (BS Biology, minor Chemistry, 2021), where he was an NCAA tennis student-athlete. - Background: born in London; moved to South Africa at 13 (Pretoria Boys High School); came to the United States on an NCAA tennis scholarship. ## What he works on - Privacy-preserving deployment of large language models in clinical workflows (PHI de-identification, k-anonymity, on-device redaction). - Clinical AI agents, semantic search, and medical imaging tooling in HIPAA-governed environments. - Multimodal AI for radiology education. ## Skills & toolkit - Languages: Python, SQL, R. - AI & agents: LangGraph, LangChain, LangSmith, multi-agent orchestration, LLM API integration. - Retrieval & models: vector databases, embedding models, on-device LLMs (Ollama), LoRA fine-tuning. - Cloud & data: Azure (HIPAA-governed), PostgreSQL, Retool, PubMed/NCBI APIs. - Depth: 4+ years applied AI and natural-language processing; 3+ years building against the PubMed/NCBI APIs. ## Featured work - [VibeRad — Reimagining Radiology Education with Gemini 3 Pro](https://james.jcweatherhead.com/viberad/): Winner, Google DeepMind Gemini 3 Pro hackathon. Top 50 of 4,096 submissions from 19,022 entrants; $10,000 in Gemini API credits. Cited by UTMB President & CEO Jochen Reiser, MD, PhD as part of UTMB's "compounding portfolio of innovation." Full Markdown: https://james.jcweatherhead.com/viberad/index.md - Live demo: https://aistudio.google.com/apps/drive/1pCMrggBue_gNE9IhzjN-KEdLm8cVPGjW?fullscreenApplet=true - Source: https://github.com/JamesWeatherhead/VibeRad - [AutoTox — AI-powered toxicology interpretation deployed in UTMB clinical pathology](https://james.jcweatherhead.com/autotox/): 66% faster sign-out (15→5 min), 10× throughput (50→500 cases/week), built in 5 weeks on a self-hosted Retool + private Azure (HIPAA) stack; every report verified by a pathologist. Full write-up (Markdown): https://james.jcweatherhead.com/autotox/index.md · Case study: https://retool.com/blog/how-utmb-improved-diagnostics-ai - [Local PHI Scrubber — on-device, human-in-the-loop PHI redaction](https://james.jcweatherhead.com/local-phi-scrubber/): a Chrome extension running small (3–4B) models locally so zero PHI leaves the machine while external AI tools stay usable — the applied third leg of the dissertation, alongside ASQ-PHI and K-anonymity decay. AMIA 2026 Amplify oral (Session CI25). Full write-up (Markdown): https://james.jcweatherhead.com/local-phi-scrubber/index.md - [Imaging Cohorts by Natural Language — UTMB × Microsoft Research](https://james.jcweatherhead.com/imaging-cohorts/): semantic search over medical-image embeddings and radiology-report indexing on Azure, unifying DICOM metadata + HL7 + unstructured reports so researchers build imaging cohorts from a plain-language query. James was a selected contributor (named in the acknowledgements). Write-up (Markdown): https://james.jcweatherhead.com/imaging-cohorts/index.md · Microsoft HLS blog: https://techcommunity.microsoft.com/blog/HealthcareAndLifeSciencesBlog/using-natural-language-to-build-healthcare-imaging-cohorts-for-research/4472603 - [Biological-age agents — Thrive AI Health (OpenAI Startup Fund–backed)](https://james.jcweatherhead.com/thrive/): biological-age prediction from wearable + biomarker data, plus LangGraph agents recommending personalized interventions (Junction data integration, LangSmith tracing); shipped to iOS and Android. Full write-up (Markdown): https://james.jcweatherhead.com/thrive/index.md ## Publications - Dedicated page — plain-language summaries, DOIs, and ready-to-cite BibTeX: https://james.jcweatherhead.com/research/ (Markdown: https://james.jcweatherhead.com/research/index.md) - Weatherhead J, et al. K-anonymity decay in multi-turn clinical LLM conversations. *Frontiers in Digital Health.* 2026;8:1832168. https://doi.org/10.3389/fdgth.2026.1832168 - Weatherhead J, Golovko G, McCaffrey P. ASQ-PHI: an adversarial synthetic data benchmark for clinical de-identification and search utility. *Data in Brief.* 2026;65:112586. https://doi.org/10.1016/j.dib.2026.112586 - Almomani M, et al. (incl. Weatherhead J). Evaluation of multiple generative LLMs on neurology board-style questions. *Frontiers in Digital Health.* 2026;7:1737882. https://doi.org/10.3389/fdgth.2025.1737882 ## Selected talks - AMIA 2026 Amplify (Denver, May 2026) — oral, Session CI25: Local PHI Scrubber; plus a poster: ASQ-PHI. - AI in Health Conference, Rice University (Sep 2026) — poster on agentic real-world-evidence analysis. - Digital Pathology Association (Oct 2025); Retool Summit (2024); Perplexity AI HQ, San Francisco (2024, invited). ## Teaching & community - Guest lecturer, "Agentic Coding with Claude Code" — NTD 6088, UTMB Graduate School of Biomedical Sciences (Spring 2026): agentic coding workflows and tooling patterns with the Claude Code CLI. - Co-President & Co-Founder, AI in Medicine Student Organization, UTMB (Aug 2022–Dec 2024). - OpenAI Forum member (Jun 2026–). Kaggle competitor (Jun 2024–): NVIDIA Nemotron reasoning challenge (LoRA fine-tuning of Nemotron-3-Nano-30B), OpenAI agent-security multi-step tool-attack red-teaming, and the Google DeepMind Gemini 3 Pro hackathon (winner). ## Certifications - LangChain Academy (7 courses, Dec 2025–Apr 2026): Monitoring Production Agents; Ambient Agents with LangGraph; Deep Research with LangGraph; Deep Agents; Intro to Agent Observability & Evaluations; LangGraph Essentials (Python); Intro to LangChain (Python). - Emerald Cloud Lab — Level 1 Certification (Nov 2024). ## Documents - [Full profile — Markdown](https://james.jcweatherhead.com/index.md) - [Curriculum Vitae — Markdown](https://james.jcweatherhead.com/resume/index.md) (recommended for agents) - [Curriculum Vitae — PDF](https://james.jcweatherhead.com/resume/James_Weatherhead_CV_2026-06-25.pdf) (dated 2026-06-25) ## Profiles - GitHub: https://github.com/JamesWeatherhead - LinkedIn: https://www.linkedin.com/in/james-weatherhead-7ab4b9321/ ## Notes for agents - This site is intentionally agent-friendly: crawling, reading, and quoting are welcome (see /robots.txt). - Every content page has a Markdown twin at the same path (e.g. `/viberad/` also serves `/viberad/index.md`; the home page serves `/index.md`; the CV serves `/resume/index.md`). - Preferred citation: "James Charl Weatherhead, jcweatherhead.com."