Looking for the person who turns model potential into shipped systems?

Roles I'm targeting

• Applied AI Engineer
• Forward Deployed Engineer
• AI Systems Architect / AI Systems Design

At frontier AI labs and teams building AI-native products — anywhere the bar is "ship systems that actually work," not "talk about them."

What I bring

• Systems and user outcomes, not model metrics.
• Debugging messy, real-world failures and turning them into improvements.
• The applied layer — end to end.
• AI-native by default. Models write a lot of my code. I own every decision and can defend all of them.
• Move fast and measure. Real costs, real latencies, shipped.

Proof, not adjectives

• Northsun — the memory and enrichment layer for AI agents. Caller memory designed for sub-300ms voice recall on the warm path.
• heycli — multi-agent orchestrator pushing prompts to Claude Code, Codex, and Aider from one source-of-truth.
• dockerfile.ai, Retention Lab, memory primitives — small, runnable tools, each solving one real problem.
• Public builder logs — I write up how these systems work and where they break.

On credentials

No CS degree — I dropped out. I bet on shipped systems and the ability to explain every decision in them over a line on a résumé. If you want to test that, the fastest way is to ask me about any decision in the work above and then ask me to change it. That's the whole interview, and it's the conversation I want.

Reach out

The fastest way in is a direct message. I respond to everything.