Get to know the problem
Follow the work, gather examples and agree what the prototype needs to do.
Over 90 days, I help your team research, design and test an AI system around a piece of work you want to improve. You learn by building it.
Bring the prototype, the results and the parts you’re struggling with. In our weekly working session, I’ll help you examine the approach, work through technical decisions and decide what to test next.
Teaching follows the project. One week we might map how information moves between departments. Another might involve comparing models, reviewing code or finding out why a seemingly good answer is wrong.
Between sessions, your team carries the work forward. We agree who will do that work and how much time they can realistically give it before we start.
Your team does the building. I help you work out the approach, review what you’ve tried and teach the skills needed for the next step.
Follow the work, gather examples and agree what the prototype needs to do.
Try the uncertain parts first. Bring what works, what breaks and your questions to our reviews.
Test it against the original problem, record the limits and work out what to do next.
We meet each week to review the work. Between sessions, your team builds and tests. Before starting, we agree who’s involved and how much time they can give it.
The aim is a prototype your team understands well enough to evaluate and improve. Any further work to put it into production is scoped separately.
Tell me about your teamYou need someone who understands and owns the business problem, plus someone able to build and test. They can be the same person. You’ll also need access to approved examples of the work.
For a dealer group, that might be an operations lead working with a developer. For an automotive software company, it could be a product manager and an engineer testing a new customer feature.
Our first conversation covers the project, your team and the help you need. If it looks like a good fit, I’ll set out the scope, schedule and fee. We also agree confidentiality, ownership, access and which tools we’ll use.
Investment is discussed on that call. Any tool subscriptions, specialist development or production integrations are identified separately in the proposal.
Want to understand the approach first? Read about Intent-to-System Engineering or look through the example automotive projects.
I’m exploring a pathway for local university students and interns to join supervised dealer-group projects. They would learn by doing, while your team gets to know their work.
Help shape the university & talent pathwayA few lines about the problem and your team are enough to start. We can talk through the rest together.