Projects

Oceanfront Hardware

A B2B checkout-first storefront, built and still maintained end to end.

Website

Took the client from offline sales to a storefront their trade customers can actually order through, as the only engineer on it.
Took the client through their own product catalogue, resolving inconsistencies that were quietly keeping products out of search results and category pages.
Server-rendered for indexability; search returns in ~20 ms, down from 700, with typo tolerance across the catalogue.

Languages:Typescript

Frontend:Next.js

Backend:StrapiMeilisearch

Services:Stripe

The Oceanfront Hardware storefront on a phone, showing a product page

Throughput benchmark: pyvolr vs fast-vollib, opengreeks, and the wider BSM ecosystem, log-log by array size on an Apple M4 Pro — Throughput vs the active BSM ecosystem (log-log, by array size)

pyvolr

A Python library for Black-Scholes-Merton and Black-76 pricing, all five Greeks, and implied volatility -- fully vectorised over NumPy and drop-in compatible with `py_vollib`/`vollib`.

GitHub

Fastest in the active BSM ecosystem at scale: 1M implied-vol solves run ~2x faster than fast-vollib's numba backend and ~12x faster than opengreeks, and 1M strikes price in ~4 ms -- on a Rust core exposed via PyO3, parallelised with rayon, and still sub-microsecond on scalar input.
Stays f64-exact down to option prices of 1e-215, where fast-vollib, opengreeks and quantforge all begin returning silently-wrong constants -- Jackel's "Let's Be Rational" solver over an `erfcx`-based normal CDF, pinned against an mpmath golden ladder and differential-tested in CI.
Wheels for Python 3.10-3.14 across Linux, macOS and Windows, including free-threaded 3.13t and 3.14t builds, published through PyPI Trusted Publishing with no stored credentials -- plus `py_vollib` and `py_vollib_vectorized` shims that migrate existing users in one import line.

Languages:RustPython

Other:PyO3NumPymaturin

Secrets Spotter

An open-source CLI and Chrome extension that detects exposed secrets in files, directories, stdin, web pages, and network traffic -- powered by a shared Rust core.

GitHub

Built a unified Rust detection engine with 60+ patterns (AWS, GitHub, Stripe, JWTs, private keys, and more) plus false-positive filtering via Shannon entropy, character-class diversity, and code-identifier rejection -- shared across the CLI and browser extension from a single core.
Shipped a CI/CD-ready CLI with SARIF output for GitHub Code Scanning, JSON mode for automation, and glob and severity filters.
Engineered a WASM-powered Chrome extension that scans DOM content and intercepts fetch, XHR, WebSocket, SSE, and cookies through a MAIN-world interceptor, with SHA-256 deduplication and automatic re-scan on SPA navigations -- all fully local.

Languages:RustJavaScript

Frontend:Chrome Extension

Other:WebAssembly

Secrets Spotter

Scanning page...

86 scanned, 0 skipped (cookie: 6, dom: 15, fetch: 65)

2 secret(s) detected

High: 1|Medium: 1

JWT Token[DOM]

eyJhbGciOiJIUzI1NiIsInR5cCI6Ikp...

JWT decode

{
  "alg": "HS256",
  "typ": "JWT"
}

{
  "sub": "1234567890",
  "name": "John Doe",
  "iat": 1516239022
}

PostHog API Key[DOM]

phc_SrqtLbDe...

Debug Log