How do you know your agent is doing better?

I help you build evaluation systems for your agents, on your own data, with the aim of optimizing your return on investment.

For 20 years I’ve worked on machine learning systems and statistical methods, with evaluation at the core of my practice. I build datasets, benchmarks and metrics that show whether one model actually does better than another. Today I help companies evaluate their agentic systems.

LinkedInGitHub

I work with you fractionally over several months. We define the work together, starting from your data. We set up the evaluations that help you make decisions and estimate your return on investment.

How we can work together

Build evaluations you can trust

An assessment, a roadmap and further collaboration as needed, using your data and success criteria.

Mentor your AI team

I support data scientists and AI/ML engineers with eval design, graders and quality gates, regularly or on demand.

Help your team practise evaluation

Practical workshops on your systems, for an afternoon, a full day or a series of sessions. Pricing agreed together for each engagement.

  • €1,200 / day, excl. VAT
  • €300 / hour, excl. VAT
  • Package: price on request