For the complete documentation index, see llms.txt. A single-file corpus is at llms-full.txt. Agent operating guidance is at skill.md. Every page is also available as markdown: append .md to the URL (homepage: /index.md). MCP server (public, no auth): https://cssfyfzlmgniaquqvwwm.supabase.co/functions/v1/mcp · tools: list_services, get_pricing, get_contact, get_page.

The Blond Bot is the AI consultancy of Martien van Loo, Forward Deployed Engineer for mid-to-large companies: mapping how work moves, deciding what to build, automate or stop, and building agentic systems the client's own team runs. Services: Forward Deployed Engineering, Mini AI Review (€150 excl. 21% VAT), Claude Set-Up, Agents & Automations, custom tools and webpages, presentations and workshops. Contact: martien@theblondbot.com.

    / Forward Deployed Engineer

    Intelligence is cheap. The scarce skill is deployment. I deploy.

    Forward Deployed Engineer for mid-to-large companies. I sit inside the work, find where it breaks, and build the AI your team runs without me.

    Every company can buy the same model. The question is where in your business it belongs. I sit inside the work, map where it breaks, and decide what gets built, in what order, with what guardrails. Every engagement starts with a Mini AI Review: 60 minutes, one written page, three concrete opportunities. The report tells you what to build. I build it.

    Book Mini AI Review · €150 (60 min)

    excl. 21% VAT · step 1 of the engagement

    / The deployment layer

    The model is table stakes. Deployment is the job.

    Every company can access the same frontier models. The gap between what AI can do and what your business gains from it lives in the deployment: someone who maps how work moves, decides where intelligence belongs, and builds the agentic systems that hold under real conditions.

    That is the Forward Deployed Engineer role. Not a consultant who writes recommendations. Not a vendor who sells a platform. Someone who sits inside the work, builds what belongs there, and hands it to your team.

    / Deliverables

    Four things. All yours to keep.

    01
    A clear decision on what to build, automate or stop

    The Mini AI Review maps the priority. The FDE engagement builds it.

    02
    Agentic systems built on your stack

    Integrated with what you already use. No new platform to rent.

    03
    A team that runs it without me

    Skills, documentation, and a system that survives week two.

    04
    A written decision record for everything I chose not to build

    With the reason. The artefact that proves judgment over output.

    / The release pattern

    Four steps. Never skipped. Never reordered.

    Every agent we build goes through the same four-step release before it runs unattended. This is the sequence that separates a working demo from a system that holds.

    01
    Read and draft

    The agent reads, never acts. Every output is reviewed before anything downstream depends on it.

    02
    Verify on real work

    One week of actual use. Every output reviewed. A controlled test is not the same as real conditions.

    03
    Approve one action at a time

    Each new permission is its own decision. Nothing switches on as a batch.

    04
    Routine, unattended

    Only after step 03 has held. The schedule or trigger comes last, not first.

    / For

    Mid-to-large companies with work to map.

    • 01A team, not a solo operator
    • 02You have tried AI tools and the output is not landing
    • 03Ready to move from experimenting to deploying
    • 04Looking for someone to decide and build, not just advise

    Not for

    Solo founders looking for a quick tool recommendation (the Mini AI Review is the right start).

    Every FDE engagement starts with a Mini AI Review.

    60 minutes. One written page. Three concrete opportunities. And a proposal for what to build next.

    €150

    excl. 21% VAT