Notes from the build
What I'm learning while building and controlling agents on real codebases.
Why Technical Founders Get Stuck (And How to Break Free)
You built v1. Users showed up. Now you're drowning in tech debt, feature requests, and architecture decisions. Here's how to get unstuck.
Build and Control Agents That Ship Real Systems
Most developers I talk to are stuck in the middle layer — past prompting, short of real agentic building. Here's how I think about closing that gap.
· 7 min read
MCPs in Plain English: Tools Your Agent Can Call
An MCP server is a list of tools your agent can call. That's it. Here's what that means in practice, with the Xano MCP I actually use as the worked example.
· 7 min read
What a Skill Actually Is (And Why It Changes How I Build)
A skill is a small, named instruction set your agent loads on demand. Here's what that looks like in my day and why it's the thing that finally made sessions stop sucking.
· 6 min read
Why Your AI Pipeline Fails 74% of the Time
Six stages at 80% reliability each gives you a 26% success rate. The fix isn't better prompts — it's fewer stages with pure functions.
· 4 min read
AI Agents Are Your New Engineering Team
You don't need to hire five engineers. You need one developer who knows how to orchestrate AI agents. Here's how the new model works.
· 3 min read
The Real Cost of Technical Debt (It's Not What You Think)
Technical debt doesn't just slow down your engineering team. It kills deals, burns out founders, and makes your product fragile at the worst possible moment.
· 3 min readI'll look at your code, live.
30 minutes. Your screen, your repo. No pitch deck. I help you solve the thing that's blocking you.