Curriculum vitae

3.9/4.0 GPA. Cross-registered at MIT, Harvard Law School, and Harvard Kennedy School.

4.0/4.0 GPA (WES equivalency). Distinction in every taught module.

4.0/4.0 GPA (WES equivalency). First in year for Health, Ethics, and Society.

4.0/4.0 GPA (WES equivalency). First Class Honours in every module. Awarded Best Student and Best Dissertation.

Awarded BlueDot Impact’s Career Transition Grant: six months of funding for full-time independent AI safety/governance research.

Designed and launched Scrutica.com to map previously unmonitored physical bottlenecks in frontier AI supply chains, providing open infrastructure for policy and governance research. Tracks 4,544 facilities in 109 countries, 21,037 supply-chain edges, and 34 sovereign programmes (July 2026), with a suite of analytical tools including chokepoint-cascade simulation and cross-jurisdiction coordination-gap analysis. Adapted platform following advice from Yafah Edelman (Head of Data and Trends at Epoch AI).

Evaluations and Collaborations Lead on a research team examining the transferability of the Financial Action Task Force governance model to the international AISI Network. Analytical contributions included classification of second-order institutional problems in AI governance (e.g. who evaluates the evaluators), analysis of threshold divergence across frontier safety frameworks, and a two-timescale governance proposal separating rapid technical iteration from slower institutional coordination. Interviewed senior stakeholders and collaborated with other research teams. Paper published on SSRN, interactive explainer microsite: thwingetal.app

Lead a research project I conceived on eval-independent governance that maps safety-critical triggers across frontier governance frameworks, marks which route through model-evaluation results, and catalogues how evaluations fail in practice (eval awareness, developer and model sandbagging, measurement-reliability breakdown) to inform a taxonomy and design principles for eval-independent triggers. Supervise a team of five (four research fellows and a research manager) with backgrounds in AI ethics (Cambridge), cognitive science (UCL), law (Sciences Po), and AI governance (the Ada Lovelace Institute and GovAI). Paper in final preparation for submission.

• Safety Under Scaffolding: How Evaluation Conditions Shape Measured Safety (arXiv:2603.10044)
• IatroBench: Pre-Registered Evidence of Iatrogenic Harm from AI Safety Measures (arXiv:2604.07709)
• Defensive AI: When Safety Alignment Creates Tort Liability for Medical Information Omission (SSRN 6402098)
• Frontier Lag: A Bibliometric Audit of Capability Misrepresentation in Academic AI Evaluation (arXiv:2605.04135)

Served on the expert Delphi panel, rating proposed practices across successive rounds alongside participants from leading labs (e.g. DeepMind, OpenAI) and orgs (e.g. AISI, METR, RAND); endorsed the resulting AI Evaluation Consensus Statement. My Safety Under Scaffolding paper was among the evidence base reviewed.

Selected for medical and legal expert cohorts and completed evals/rubric training; visa restrictions precluded taking paid evaluation work.

Validated the clinical accuracy of AI-generated medical consultation reports (preceded current work in AI safety).

Competitive international fellowship; full funding for graduate study at Harvard (tuition, health insurance, and living stipend, ~$150k).

Awarded the £2,000 first prize (and publication in Sweet & Maxwell’s Tort Law) for an essay challenging the legal distinction between physical and psychiatric harm, judged by The Honourable Mr Justice Cotter.

Perfect score (170 verbal, 170 quantitative).

Team of four awarded for multimedia teaching on practising within one’s competence.

Team of two awarded second place nationally for educational video content on diabetes.

Doubleday Prize for essay on interpersonal skills in medicine. Presented at the RSM and published in Lifestyle Medicine (Wiley).

Essay on the ethics of AI in medicine.

Highest overall mark and highest dissertation mark (mixed-methods project analysing British media coverage of COVID-19 lockdowns).

Highest mark of 200+ medical students in the core Health, Ethics, and Society module (96% average).

• Fuller Futures: Strategy and AI integration for a new non-profit connecting resources to frontline high-impact charities. Thesis: agentic tooling lets small mission-driven teams operate at institutional scale. Founding team of leaders from the WHO, NHS, and Parliament.
• Progenic Genomics: Advising on policy in a domain (accessible embryo health screening) where technology has outpaced regulation.
• SEAL: Policy briefs on maternal and child health, affordability, and healthcare reform for NJ gubernatorial campaign.

Designed and ran a research survey across 30+ UK medical schools (500+ responses) on ethics, public-health, and lifestyle-medicine training; three manuscripts in peer review. Led a core team of six, and coordinated with 30+ medical school representatives.

• EHC: Led analysis and write-up of a five-year NSTEMI service evaluation; co-authored British Journal of Cardiology paper.
• UI: Led quality appraisal for systematic review on smoking-cessation interventions; co-authored European Journal of Public Health paper.

Selected from 3,000+ applicants for a clinician-healthtech fellowship.

Ran the medical ethics conference team. Organised two national conferences (100+ attendees at each). Led stage interviews.

Tort, clinical negligence, and health policy work; legal research and case prep; hearings at the Royal Courts of Justice.

Completed internship at London, Oxford & Brighton offices; worked across IP law matters.

Part of the founding committee of Edinburgh’s branch of Raise. Raised ~$25k for AMF and spread awareness about GWWC pledge.