About

Workflow improvement consultant. AI implementation engineer.

I work with solo founders and micro-corporations who want their daily operations to run on systems instead of willpower. The work is grounded in listening — to operations, to constraints, to the parts that already work — and ends with a system that runs without ceremony.

Stance

Augmentation, not replacement.

I do not replace people. I extend what is already working — quietly, in the background, on shifts the founder no longer has to keep. The system has to leave room for the operator to refine the product, the offer, and the relationships. That margin is the entire point.

Pre-engineering layer

I read the workflow before the code.

Most automation work fails not in the implementation but in the framing. Before any code is written, I sit with the operations, find where automation pays back, and find where the human has to remain. The deliverable is a workflow first — code is the artifact at the end of that thinking, not the start.

Engineering

I run the technical surface — you do not.

n8n, API integrations, LLM orchestration, multi-tenant data isolation, e2e tests — that whole technical surface lives on my side. The handoff is a workflow that runs and operating documentation that explains what is happening, in your language. You do not learn a new admin screen unless absolutely necessary.

Background

Non-engineer to AI implementer.

I came up as a non-engineer. The framing of the problem stays in the operator’s language because that is the language I think in. The technology — Claude Code, n8n, OpenAI / Anthropic APIs, Supabase, Playwright — is in service of that framing, never the other way around.

Working relationship

External member, not consultant.

No 100% complete spec, no fancy documents required. We look at rough notes and the screens you actually use, and we ask together: which of these does a human still need to be in the middle of? Start small, stop if it does not fit. That posture is the entire offer.

Bring the workflow you would describe to a new hire.

Thirty minutes is enough to identify where automation pays back, and to decide together what to build first.