I work on high-order numerical methods for fluid flow at Oxford, and on machine-learning systems for quantitative venture investing.
I'm a DPhil student in Engineering Science at Oxford, sponsored by Dstl, working on computational fluid dynamics: an arbitrary-order discontinuous Galerkin solver for the 3D Euler equations, and a generalised eigenvalue formulation of unsteady potential flow that identifies efficient swimming kinematics without searching over them. Before Oxford I was a naval architect at the Ministry of Defence.
Alongside the doctorate I'm an AI research scientist at Vela Partners, where the problem is rare-event prediction with almost no labels: which early-stage companies become outliers, and how many bets a fund has to make before that question stops deciding its returns. The work has produced a preprint, a patent-pending classification method, and the internal tooling the research runs on.
Some things here are neither: software built because the thing should exist and didn't.
Selected work
High-efficiency swimming from hydrodynamic eigenmodes
2025PaperEfficient swimming kinematics found by solving a generalised eigenvalue problem, rather than by searching the space of motions.
Fortran · Python · BLAS / LAPACK
R.A.I.S.E.: Reasoning-Based AI for Startup Evaluation
2025PaperA memory-augmented, multi-step LLM framework for predicting which early-stage companies become outliers, a rare-event problem with almost no labels.
Python · LLMs · Rare-event evaluation
Reasoned Rule Mining
2025PaperA two-stage, precision-optimised LLM classifier for quant VC: 24.5% precision against a 2% base rate, a 12.3x uplift.
Python · LLMs · Calibration
AutoAlbum
2026LiveA stamp album page designer that exports true 1:1 vector PDFs. Millimetre-precise, entirely in the browser, nothing uploaded anywhere.
Next.js · TypeScript · jsPDF · Tailwind
Elsewhere
Degrees, papers, and the rest of it are on the CV. For anything else, email is best.