You take AI from prototype to production. On data you can trust. Most AI projects don’t fail on the model, but on the data underneath. At Beeminds, that data is already in order. With the Intelligenthive®, our AI data platform on Microsoft Fabric, we build a reliable data foundation within weeks for organisations such as UNICEF Netherlands, SportCity, Swinkels and Samenwerkende Kinderopvang. On top of it, you build the AI applications people use every day: agents, assistants and automated workflows.
In this role you work on both our product and client implementations, with a small team of data engineers, data consultants and architects, for several clients at once. This isn’t a solo secondment: you build on and with our own product. You keep up with the latest in AI, go to events and turn what you learn into reusable building blocks in our platform.
About the role
As an AI Engineer, you’re the link between the data in the Intelligenthive® and the work people do with it every day. You build AI agents and multi-agent workflows that answer questions, carry out tasks and prepare decisions, based on the data in the Intelligenthive®. You design RAG solutions on structured and unstructured data: semantic models, documents and source systems such as SAP. You work in Azure AI Foundry and with Azure OpenAI, and for each question you pick the model that fits best on quality, speed and cost. Before anything goes live, you set up evaluation with test sets, metrics and monitoring: you know how good your agent is, and you can show it. You take solutions to production with Git, CI/CD and containers, keep them running and take care of the essentials: access rights, logging, cost control and the requirements of the GDPR and the EU AI Act. You build reusable components into our product, so the next client goes live faster. You sit at the table with the client yourself: you ask which work should disappear, and you say so when AI isn’t the answer.
What do you bring to us?
- You build in Python, your code is readable for the next engineer, and you know DevOps: Git, CI/CD, Docker and infrastructure as code.
- You’ve taken LLM applications to production: RAG, prompt engineering, tool calling and vector search aren’t theory to you.
- You know at least one agent framework, such as Semantic Kernel, LangGraph or AutoGen, and you also know when you don’t need one.
- You measure instead of hoping: you set up evaluations and know why an agent fails.
- You think in data: you understand how a lakehouse and a semantic model work, and why an agent on bad data gives bad answers.
- You have a few years of experience as an AI engineer, ML engineer, data scientist or software engineer, and you can explain a technical choice to a director without jargon. Bachelor’s or master’s level (HBO/WO).
- Nice to have: C#, Microsoft Fabric, Azure AI Foundry or Copilot Studio, classic ML (scikit-learn, PyTorch), MCP and experience with the EU AI Act or responsible AI.
- You speak and write fluent Dutch and good English.
What do we offer?
-
Salary and holidays
A good salary that matches your knowledge and experience, plus paid holidays.
-
Mobility budget
Your choice: an NS Business Card (Dutch rail pass), mileage allowance, e-bike or lease car.
-
Your choice of devices
Pick your own laptop and mobile device. A new device every year through device-as-a-service, with the right specs.
-
Time to learn
Part of your working time is for self-study, side projects and knowledge sharing. We run hackathons and bring in speakers.
-
Bee in Control budget
€5,000 a year for training, (international) conferences, events or a SaaS tool you want to try out. No complicated approvals.
-
Open culture
No politics: everyone is free to say anything. Our core values BeeTogether, BeeHonest and BeeDifferent are more than a poster on the wall.
-
Flexible and hybrid
Flexible hours: run your errands, skip the traffic and properly switch off on holiday. Work from home, in Naarden or Heerlen, at the client or now and then from abroad.
-
Driven colleagues
Colleagues who challenge you, support you and value good work. In an informal atmosphere, with team outings, lunches, knowledge sessions, drinks and events.
The application "process"
- A quick “call”By phone, on Teams or at the office.
- Vision and ambitionWe talk about our direction and your ambitions.
- Team and workYou meet your colleagues and see the work.
- The offerBoth excited? Then we make you an offer.
