Welcome to Chipp Engineering
The Chipp engineering blog — deep dives on how we build the platform for AI builders, the decisions behind it, and what we learn along the way.
This is the Chipp engineering blog. Expect long-form, honest writing about how we build the platform — the systems, the decisions, the things that broke and what we did about them.
What you’ll find here
- Architecture deep-dives. How our Alchemist, tool-execution pipeline, vector search, voice stack, and billing system actually work.
- Postmortems. What broke, why, and how we fixed it. Public so other teams building AI platforms can skip the same mistakes.
- Performance and scale. Work we’re doing to keep p95 latency tight as we ship 20–30 deploys a day to production.
- Decisions. Why we picked Deno, why we wrote our own Stripe token billing layer, why we use a Cloudflare Worker as an edge proxy in front of everything.
Why a separate blog
Our docs are reference material — how to use the platform. The Ghost blog is for product news and announcements. This is different: it’s the engineering reasoning behind the product, written for people who build things.
If that’s useful to you, the RSS feed is the easiest way to follow along.