Training is most useful when people can connect it to the work in front of them. I shape workshops around your team’s roles, current experience, and goals, combining practical exercises with the engineering judgment needed to choose, use, and evaluate AI systems.
Start at the right level
We agree the audience and learning goals before choosing the material. A product team exploring new workflows needs a different session from engineers building tool integrations. Exercises reflect the team’s starting point and the decisions they need to make.
Learn through a relevant task
Sessions can cover using AI tools, preparing useful context, designing agentic workflows, evaluating outputs, or understanding the infrastructure around agents. Participants work through examples and examine where a system needs clearer instructions, better evidence, or human review.
Connect learning to implementation
The session ends with a practical next step: an approach to try, a workflow to improve, or a set of evaluation questions to use. Training can stand alone or accompany a scoped build so the people using and maintaining the system understand how it works.
What we scope together
Deliverables are agreed around your needs. A typical scope can include:
An agreed workshop plan
Topics, audience, learning goals, and exercises shaped around the team’s work.
Practical working material
Examples, exercises, and reference notes that support the session.
A path to continued practice
Concrete next steps the team can use to apply and develop what they have learned.
Before we begin
Is this only for engineers?
No. Workshops can be shaped for product, business, or engineering teams. We agree the technical depth and learning goals for the people attending.
Can training accompany an implementation?
Yes. Training can use the workflow being built to explain how it operates, how to review its outputs, and how the team can improve it.
Can we use our own workflow as an exercise?
Yes, where suitable. We agree the material and access needed in advance, using sanitized examples when the original work contains sensitive information.