- Salary
- €4k – €6k
- Location
- Den Haag, Zuid-Holland, Nederland
- Workplace
- Hybrid
- Type
- Full-time
- Department
- Product & Technology
- Seniority
- Senior
- Source
- Recruitee
Description
Legal AI is only as good as the knowledge behind it. At Lefebvre Sdu, we’ve been building that knowledge for 200 years. Day in, day out. Law by law. Word by word. Other legal AI tools run on our data and software. As the Dutch heart of Lefebvre Group, the European leader in legal and tax knowledge, we’re also developing our own platform: GenIA-L, where lawyers and tax professionals don’t just get an answer, but an answer they’re willing to put their name to. More than 75,000 professionals use it every day, accounting for over €100 million ARR across Europe. We’re looking for a Medior AI Engineer to help shape how the intelligent core of that platform evolves.
HOW YOU'LL MOVE US FORWARD
GenIA-L is a research platform for legal professionals. A user asks a question. The platform searches our trusted sources and turns them into a useful answer. It does this with agents: specialised AI components that each handle part of the question and work together. In the legal domain, an answer can't just sound fluent. It has to be correct, and you must be able to see which source it comes from. This is not a pilot project: the platform runs in production and is used heavily every day.
As a medior, your focus is on building. You work independently, on challenges that come from customers and on challenges we spot ourselves. You break down a legal process, translate it into a technical process and work out how to best connect the pieces. Stuck on something? You say so, and your colleagues think along with you.
You build and improve the agents of GenIA-L and the way they work together.
You build and maintain our knowledge packages: for each legal article, an expert-validated set of commentary, case law, parliamentary history and practical interpretation. You also work on the pipelines that create these packages and keep them up to date.
You make quality measurable. With evaluation sets, observability and metrics, you show that answers, source attribution and reliability really improve. Not by gut feeling, but with evidence.
You build features for customers, such as analysing their own documents and new ways to search case law.
You work directly with editors and subject-matter experts, and check with them which parts of a legal question really matter.
You come up with and test ideas to make the product sharper or faster, and alternate AI experiments with back-end development.