How I work

Start small. Prove value. Then build.

Each step is fixed in scope and stands on its own. You only move to the next one when the last one has earned it.

  1. 2–3 weeks

    Diagnose

    I work with your people and your data to turn a vague question into a clear problem, requirements and a solution architecture.

    You get: a sharp problem definition, requirements, a solution architecture, and a plan to build it.

  2. 4–6 weeks

    Prove

    I build a working proof of concept on your own data and validate it honestly, so you know whether it works before you invest.

    You get: a tested prototype, the evidence behind it, and a clear go or no-go.

  3. Ongoing

    Build

    I lead delivery as your fractional technical product lead, with specialists from my network or your own team.

    You get: a production solution, and a team that understands why it was built this way.

Problems I solve

The questions behind most data projects

We need to know why it happened.

Root-cause systems that detect what changed, find what caused it, and recommend what to do.

Our AI agents demo well. Can we trust them?

Evaluation pipelines that test agents before launch and keep checking them after.

Our data project is stuck between teams.

One end-to-end design that data science, engineering, analytics and product can all build against.

Which customers need attention?

Churn and recommendation models tied to the actions your teams can actually take.

Where does AI actually help?

A sober look at which of your processes benefit from AI, and which don’t.

Something else?

The best projects often start with a question that doesn’t fit a category.

Tell me about it →
Questions

Before we talk

Who is this for?
Mid-sized companies with a data or AI initiative that’s still vague, stuck between teams, or missing senior technical product leadership.
Do I have to commit to all three steps?
No. Each step is fixed in scope and stands on its own. Many engagements end after Diagnose with a clear plan your own team can run with.
What does it cost?
Diagnose is a fixed fee, agreed before we start. Prove and Build are scoped and priced once we know what’s worth building.
Do you need access to our data?
For Prove, yes. We agree up front on what access is needed, where the work happens, and how data is handled.

Have a problem worth untangling?

Book a free 30-minute call and tell me what you’re working on.

Book a 30-minute call