6+ years building digital technology products, at various scales, and in various industries, from fintech to conversational AI.
Based in Texas, USA.
I make slow things fast, teach whatever I know to anyone who'll listen, and spend too many weekends in a racing sim.
I was the kid in the neighbourhood everyone called when a computer misbehaved. Then, in the first year of my maths degree, I turned some famous formulas and theorems into algorithms, and I never really stopped coding after that. I love the craft of it.
These days I build web products that are fast and hard to break, from fintech to conversational AI. Performance is my favourite topic, enough that I once talked about it on stage at Google DevFest in Glasgow.
I've spent years teaching maths and computer science too, most recently Probability for AI at Stanford's Code in Place. Sharing what I know is one of the most fulfilling parts of the job.
Away from the keyboard I'm an F1 and WEC fan with a sim-racing rig of my own. I came to Texas for grad school and stayed. I like it here, but I'm open to wherever the next good problem is.
Independent · Remote
AHRO Research Centre · Remote
Raenest · Remote
eBanqo Inc · Remote
Stutern · Remote
Code Cradle · Remote
Lamar University · Texas, USA
Research: exponentially fitted Euler–Maruyama methods for stochastic differential equations (SDEs).
Lagos State University · Lagos, Nigeria
Thesis: numerical methods, Runge-Kutta and linear multistep methods.
Freelance work for a bakery that supplies supermarkets. Stores place orders, the bakery runs them, orders only count once payment clears, and stores get 48 hours to change their minds.
The public face of the conversational AI platform I spent four years helping build, with a hero chat animation I hand-coded.
A freelance build with one rule: beautiful, animated, and still 100/100 on PageSpeed Insights. Next.js, GSAP and MDX, so editors never need a deploy.
A CLI I wrote so nobody has to copy WordPress posts into Sanity by hand. It maps schemas, moves metadata and re-uploads every image.
Co-authored the Next.js PWA starter so nobody has to fight service workers alone: offline support, push notifications, accessibility defaults and CI/CD. About 950 stars and 200+ forks later, it's still going.
Eight months building the money parts of a fintech app: multi-currency invoicing, virtual cards, international wires and FX. I also rebuilt auth, which cut support tickets by 65%.
PCA (8 components, 84% of variance) and feedforward neural networks classifying U.S. residential properties into price classes. Low and Medium are easy; High and Very High are the hard ones.
SvelteKit had no next-themes, so I ported it. System preference detection, SSR-safe persistence and per-route forced themes.
A SvelteKit blog fed by headless WordPress, so editors keep the tools they know. Deployed on Netlify and covered by Playwright end-to-end tests.
RBF network with k-means-seeded centers, classifying a synthetic tumor microenvironment dataset with an enclosed, imbalanced class geometry. Best model: 0.9407 test macro-F1, 0.9973 ROC-AUC.
Got a freelance project, a full-time role, or just a question about software development (or math, or even motorsports)? I'd love to hear from you.