Work with me

Three formats, all built on the same idea: your team gets good at AI on your own stack, and keeps shipping after I leave. Every engagement has a fixed scope, a visible price, and an end date.

  1. Eigenwise Workshop

    One day, on site or remote

    For: Teams that want a serious, hands-on introduction to working with AI on their own stack.

    Outcome: Your team leaves with working setups on their own machines and one or two of your real use cases solved during the day.

    • A preparation day on my side: I study your stack, your tools, and your use cases first
    • Hands-on sessions built on your codebase and your standard procedures
    • Claude Code, skills, and MCP fundamentals for engineers
    • A written recap so the material survives the day

    from €2,900

  2. Eigenwise Sprint

    Six working days, spread over about four weeks

    For: Companies that pay for AI tools and want measurable output from them.

    Outcome: Three weeks later your team ships with AI and you can point at the exact workflows that got faster.

    • Audit: I name the 3 to 5 workflows where your team loses the most time and set a baseline
    • Build: a proper Claude setup for your company, custom skills for your procedures, and up to 2 standard-complexity MCP integrations for your internal tools
    • Enable: hands-on training per team, engineering first, plus a written playbook you keep
    • A prerequisites checklist before day 1 so the sprint starts on time
    • Check-ins at 30 and 60 days after delivery to make sure the new way of working stuck

    from €8,500

  3. Eigenwise Advisory

    Ongoing, about 2 days per month

    For: Companies that want a standing AI brain at the table.

    Outcome: Architecture reviews, build-versus-buy calls, hiring help, and an honest opinion when a vendor promises magic.

    Limited to 2 clients at a time. As of June 2026, both slots are open.

Past work

Contractify

Contractify, a Belgian SaaS company, built their first AI features on LangChain and found them hard to maintain and extend. I replaced that foundation with Atomic Agents and built Ada, their data-extraction agent. It runs in production today, and I still work with their team on new AI functionality.

Henkel

I designed and delivered an MCP workshop for a team at Henkel: what the protocol does, where it fits, and how to build servers for internal tools.

Start here

Tell me in two lines what you are working on and where AI is letting you down.

Book an intro call