Lead Data Scientist

Role: Lead Data Scientist

Location: Berlin

Working Arrangement: On-site full-time for the first six months

Salary: Up to €120k, 10% performance bonus, and equity available for standout candidates

You know LLMs beyond calling an API.

You’ve worked hands-on with language models, taken them beyond the experimental stage and dealt with the difficult parts that come with getting them into production. Evaluation, unreliable outputs, messy unstructured data, model behaviour and building systems that people can actually depend on.

Now you’re looking for the chance to take real ownership of how that work is done.

We’re hiring a Lead Data Scientist to lead a small Data Science and AI team working with a genuinely unusual dataset: two years of real clinical conversations.

Language is at the heart of the problem, so LLM experience is an important part of this role. You’ll be working on problems around understanding and extracting value from complex conversational data, deciding where LLMs are the right answer, where they aren’t, and how to build reliable systems around them.

This isn’t an LLM role where everything stops at prompting or a proof of concept. You’ll need to think about how models are evaluated, how outputs are tested and improved, how the underlying data and pipelines are built, and what it takes to put something into production that the wider product can rely on.

You’ll work directly with the Chief AI Officer, leading three Data Scientists and AI Engineers while remaining very hands-on yourself. You’ll set technical standards, challenge existing approaches and help decide the architectural direction of the Data Science and AI function.

Within your first 90 days, you’ll get deep into the existing data architecture, understand the models and pipelines already in place, establish how the team works and ship something meaningful into production.

We’re looking for someone who has built and shipped LLM or NLP-based systems, rather than someone whose experience is mainly theoretical or experimental. You should be comfortable working with large, messy and ambiguous datasets and building the pipelines that your models depend on.

You’ll probably have worked in a startup or similarly fast-moving environment before. You’re comfortable when there isn’t a ready-made answer, you can reason through difficult technical decisions from first principles and you’re happy defending those decisions when challenged.

As the team grows, you’ll have a say in hiring, architecture and how the wider Data Science function develops. But this is not a hands-off management position. You’ll be expected to code, build, experiment and solve problems alongside your team.

You’ll be based in Berlin and working on-site full-time for the first six months. There will be more flexibility after that, but this is a team that values working closely together in person.

There’s also a 10% performance bonus, with equity available for standout candidates. The wider team is international, with around 30 people from more than 16 nationalities, and the business is already working with some of the world’s largest pharmaceutical companies.

If you’ve got serious hands-on LLM experience, enjoy difficult data problems and want the opportunity to shape both the technology and the team around it, please apply now or get in touch to find out more.

We welcome diverse applicants and are dedicated to treating all applicants with dignity and respect, regardless of background.

All applicants are screened by an experienced, human recruiter – not AI.