I build small, sharp tools.

cynix is a one-person operation making data extraction and monitoring actually usable. Most of what I ship runs on Apify — actors that turn messy public sources into clean, typed records you can drop into a pipeline without writing a scraper.

What I care about

• Public data, clean interfaces. SEC filings, arXiv, FX rates, job boards, recipes — the data is free. The ergonomics are the product.
• No keyhole APIs. Where possible I build on keyless, openly-licensed sources. You shouldn't need an account to get yesterday's EUR/USD.
• Boring architecture that works. Snapshot-and-diff. Schema fingerprinting. Chunk-on-headings. The unglamorous patterns that don't page you at 3am.

Off the clock

This site is also a notebook. Some posts here are about the actors; others are just things I'm thinking through — reverse-engineering a rate limit, a terrible parsing bug, a tool that almost worked. The README is the product; the blog is the process.

Stack

TanStack-ish static site, hand-written CSS, deployed to Cloudflare Pages. No framework, no build step, no tracking. If it's slow, that's on me, not on a 4MB JS bundle.