AI Engineer
You've built and validated your own models. You know how to take an AI workflow from idea to something that holds up under scrutiny. Perhaps you've got a soft spot for psychology too, or at least a curiosity about the human mind. Now you want to work on something that actually matters.
Our platform already supports mental health professionals with session reports and documentation, and we're growing into diagnostics and treatment. Every week, new care organizations join.
What you'll do
What do you bring?
What we're building
At Ontzorgd we build AI software for mental healthcare: from automatic session reports to a platform that gives clinicians meaningful insight into their clients' progress. Today we save clinicians 3–4 hours of admin time a week; next, we want to help improve the care itself. Think of surfacing blind spots a clinician might miss, sparring on hypotheses about what's driving a client's symptoms, tracking whether a treatment is actually working, or mapping how complaints, risk and protective factors hang together in a network.
Where we stand right now
We've already built the data architecture for our research projects and are starting to collect data. Two research lines are taking shape, both for adults with an autism spectrum disorder: EAGLE, a collaborative R&D project on recognizing symptoms from audio and text, and Narrative ROM, where we rethink outcome monitoring based on session narratives instead of questionnaires. Both are backed by research grants, so there's room to do this properly in terms of scientific and technical aspects.
A day in the life
Reading recent AI papers, thinking through approaches, and working out how mental health constructs like symptoms, risk factors or therapeutic alliance can be captured with today's models.
Turning the most promising idea into an experiment, running evaluations, and extending the pipelines that tell us whether an approach actually holds up.
A developer on how to productionize something, or a psychologist on whether the model's output means what we think it means.
Talking with research partners, universities, and the mental health institutions we work with, aligning on studies, data and what to validate next.
Learning new things through relevant trainings, courses or the occasional conference.
Team getaways, days out on the water and other plans that have nothing to do with benchmarks.
Okay. Maybe not every day. It would be a very long day if you do all of these at once.
What do we offer?
Gross per month, competitive and depending on your experience.
The application process
An informal conversation about who you are and whether it clicks.
We dig into a model or AI workflow you validated, and you meet the people you'll work with.
A practical research and implementation case: how would you tackle it?
Does it click on both sides? Then we make you an offer.
Ready to join?
Send your CV or LinkedIn profile to sander@ontzorgd.ai. Tell us briefly about a model or AI workflow you validated from start to finish, and if you have an affinity or experience with psychology, mental health or healthcare in general, we'd love to hear about that too. We care more about how you think than where you've worked.
.svg)
