Chuck Reynolds

Chuck Reynolds

AI data infrastructure & enterprise API platforms. Developer experience. Twenty-six years of making the web actually work.

Hi, I'm Chuck. I'm a Staff Product Manager, Growth at Wikimedia Enterprise, the commercial API platform for Wikipedia, Wikidata, and the rest of the Wikimedia projects. We build the delivery mechanisms: the APIs, pipelines, and tooling that let commercial reusers consume it at scale and at high frequency. Top AI labs and search engines use it two ways: as bulk data for training, and as live retrieval and grounding at inference. Tens of millions of articles across hundreds of languages, one of the largest open knowledge graphs on the internet, and a real-time firehose of every edit as it happens.

The work I'm best at is developer experience and the data plumbing underneath it. CLI tooling, API specs and docs, and pipelines that turn messy structure into something usable on day one.

I build with agents as much as I build for them. Multi-agent workflows in large codebases: continuous code improvement, review of new work, security scanning across the whole repo. Micro models for intent interpretation. Planning sessions that used to take a team several weeks. Specs, docs, evals, data wrangling, one-off scripts. Terminal and browser mostly, less IDE than when I was writing every line myself.

It works because writing a good spec for an agent is the same job as writing a good PRD, which I've been doing for years, and because I stay in the loop to course correct. That's the part people skip.

I build the groundwork too, so a team can do this safely: agent instruction files, a knowledge hub, reusable skills for growth and marketing, tone and disclosure rules for anything published, and tooling for keeping PII and secrets out of agent workflows. Guardrails before the team needs them, not after something goes wrong.

Twenty-six years got me here. Software engineering throughout, applied to websites, ecommerce, marketing applications, and predictive analytics at Levers: ML for revenue outcomes, back when we just called it modeling. Years of it inside CBS Interactive, later ViacomCBS, as a Senior Technical SEO Analyst across CNET, ZDNet, Gamespot, TechRepublic, MetaCritic and others, at a scale where mistakes are expensive. Red Ventures bought the CNET Media Group side of it in 2020.

Two decades of how information gets structured, ranked, retrieved, and served. That turns out to be the same problem AI is solving now.

What I'm building

Previously

Talk to me

email: