AI systems architect
I build and operate production AI systems.
I build agentic systems that act on real business data, then I keep them running: the control flow, approval gates, idempotency, and audit trails that make autonomous work safe to trust. I partner with teams who own sales and front-line support. I own the engineering, the features, and the fixes.
Selected writing
All writing- The Most Important Table Is the One That Does Nothing
The smartest thing an anomaly detector can do is nothing, in the sense that it never touches a production account itself. It hands the finding to a table, and something else, under its own policy, decides whether to act.
- I Almost Chose My Agent Framework for the Job Market, Not My Clients
For months I wrote about 'an agent framework' instead of naming one. That wasn't a style choice. It was cover for a decision I hadn't made: LangGraph, Google ADK, or PydanticAI, and which career each one actually bet on.
- Let an Agent Ship Work While You Sleep Without Letting It Decide Anything
The design question that came out of a crashed orchestrator: how do you let an agent run unattended without ever giving it a decision to make? A closed decision table, four failed review rounds, and the line between orchestration and judgment.
What I build
- Agentic workflows on real business data
- Human-in-the-loop approval & resumable runs
- Framework-free orchestration cores
- Idempotency & audit-trail design
- LLM evaluation & regression harnesses
- Retrieval & data pipelines
Work with me
- Get in touchSend a message
- PartnershipsI build and run these systems for a small set of partners
- Elsewhere GitHub · LinkedIn