Bridging Business Software and AI
David Golverdingen
Senior Engineer & MCP Architect at Warmtebouw
- Agentic AI Architecture
- MCP & Agentic Interfaces
- Forward Deployed AI Engineering
- Enterprise AI Enablement
The most valuable AI work isn't about the models. It's the systems on either side of them: the business data an agent can reach, and an engineering process that holds up when the agent writes the code.
With 18+ years in software engineering, I now lead the development team at Warmtebouw with an AI-first approach, both in what we build and how we build it. Day to day I work as an internal forward-deployed AI engineer, alongside the users and domain experts whose problems the systems are meant to solve. That runs from MCP servers and ML pipelines to full-stack applications built with AI-assisted development, and what I am after in all of it is the point where AI stops being a demo and starts being something colleagues rely on.
Career
- Years Experience
- 18+
- Industries
- 7
At Warmtebouw
- Production MCP Servers
- 16
- Active Business Users
- 105
As of September 2026
Organizations I've worked with
Coming up
Career
Senior Engineer & MCP Architect
Warmtebouw
Senior Frontend Developer
Pro-Fa Automation
Senior Full Stack Developer
NS (Nederlandse Spoorwegen)
Senior Frontend Developer
Technolution (Rijkswaterstaat)
Senior Frontend Developer
ASN Bank (de Volksbank)
Senior Frontend Developer
KLM Royal Dutch Airlines
Lead Frontend Developer
Allianz
Bolster Safety 2018–2020 Softpak BV 2017–2018 Magion 2013–2017
Building software since 2007
The Missing Layer
Introducing Introspective Context Engineering for MCP: a method for building Rich Domain MCP Servers where the AI doesn't just query your data, it understands what it means. Based on multiple MCP servers running on real production business data.
Latest Insights
What Is an MCP Server? Sixteen in Production for Agentic AI
An agent that can reach your data still does not understand it. What an MCP server is, and what sixteen of them in production taught me.
2026-09-11 · 13 min readDon't Generate the UI. Let AI Compose It.
The tempting way to give an agent a real interface is to let it write the frontend. The better way is to give it a small set of trusted primitives and let it choose. A form, a chart, a map and a table, selected at runtime around the conversation.
2026-09-09 · 8 min readScale Capabilities Before You Scale Agents
Sixteen MCP servers, 103 tools, and no network of agents anywhere in the estate. Why the count that matters is capabilities, not agents.
Two ways in
Sixteen production servers and 103 tools at one mid-market firm, and what it takes to run them.
Production MCP: A Practitioner's Guide 9 parts The engineering systemHow the work gets planned and reviewed when agents write most of the code.
We Replaced Jira With Markdown Files 3 parts