We’re starting a blog. Not because the world needs another company blog, but because the work we’re doing at Purple AI deserves a permanent home — somewhere we can think out loud, share the choices behind p0, and tell the stories of teams shipping features faster than they ever have.
This is the first post. The format is intentionally simple: Markdown, a static site, no clever distractions. The same way we want the product to feel — fast, honest, and out of your way.
What you’ll find here
We’re going to keep things tightly scoped to four kinds of writing. Each post will live under one of these:
- Product — what we shipped, why we shipped it that way, and what we learned from the rollout.
- Company — milestones, hiring, fundraising, point-of-view essays. The kind of thing where we want to be on record.
- Engineering — deeper technical notes from the team. Architecture choices, tradeoffs, and the occasional fenced code block when it earns its place.
- Customer Stories — what real teams built with p0, told as honestly as the customer is willing to let us tell it.
If a post doesn’t cleanly belong in one of those buckets, it probably shouldn’t be here. We’d rather post less and mean it.
Why this matters now
Coding agents went from novelty to infrastructure in less time than it takes most teams to plan a quarter. The primitives — Claude Code, Cursor, MCP servers, plan mode — already work. What’s missing is the workflow: the layer that turns a rough idea into a reviewed pull request without a human shepherding every step.
That’s what p0 is for. And if we’re going to ask engineering teams to trust an autonomous workflow with their codebase, we owe them transparency about how it works, where it’s going, and what we got wrong on the way there.
So: welcome. Subscribe to the RSS feed if that’s how you read things. Or don’t — the posts will still be here when you need them.
— Christian
PS: We also got new swag today — Mario, Christian, and Andres in the picture above showed them off in our office in the House of AI today.