Develop and deploy machine learning models trained on large-scale medical data, improving their predictive performance, calibration, and robustness.
Contribute to statistical modelling for capacity management: combining and extending prediction models (e.g., admission inflow, discharge, survival models) into unit- and hospital-level insights, with attention to uncertainty, calibration, and simulation of patient flow.
Be involved in all stages of the ML process for new products: from problem definition and algorithmic approach through analysis, pipelining, experimentation, interpretation, certification, and implementation, all the way to post-market monitoring and reporting.
Dive into heterogeneous healthcare datasets to improve data quality and model reliability, while building scalable, generic solutions applicable across hospitals.
Write professional, maintainable, well-tested code following best practices for medical applications.
Contribute to R&D and bring new techniques and ideas into the team.
Use AI tooling wisely and responsibly to assist in work.
Participate in validation and verification activities, according to the latest medical and information security regulations.
Be present in the office a couple of days per week.

A Master's degree (or equivalent demonstrated experience) in Data Science, Computer Science, Statistics, or a similar field
Strong proficiency in Python, including commonly used libraries (e.g., scikit-learn, pandas/polars, numpy)
Solid knowledge of the entire machine learning lifecycle — from data exploration and preprocessing to model experimentation, development, validation, deployment, and monitoring
Energy, curiosity, a product mindset and a learner's mindset; a good team player
Fluency in written and spoken English
Eligible to work in the Netherlands (EU). Unfortunately we are not able to sponsor visas at the moment.

Stock Appreciation Rights (SAR) program
Development Day once a month – spend 5% of work time for personal development
Regular social events, with quarterly outings and yearly off-site
Sponsored contribution of an equivalent of 6% of salary to a private pension fund
Extra paid additional birth leave and maternity leave
Flexible working arrangements in working hours and working from abroad
25 paid holidays per year based on full time employment
Option to buy 5 extra holidays based on full time employment
Possibility to take a 3-months sabbatical once employed with a permanent contract
Lunch provided daily at the office, with plenty of vegetarian and vegan options
ClassPass and OpenUp subscription
Option to customize public holidays: swap some standard public holidays for alternative days to accommodate personal or religious observances
Company laptop, phone allowance and home office equipment provided

Hybrid — Amsterdam

Introductory chat: 15-30 mins online with one of the engineers or data scientists to align on fundamentals, salary expectations, motivation, and main requirements.
First Interview: 45 minutes online or in Amsterdam office with two colleagues from the Product-Tech Team to get to know each other, discuss background, experience, motivation, and fit. May be combined with step 3.
Technical Interview: 1.5 hours in person at the office (or online if needed) with two data scientists. Focus on machine learning lifecycle, statistical modelling, code quality, and solving realistic modelling problems.
Final chat (+ optional lunch at the office): 45 minutes with the CTO and/or Head of Engineering and Data Science to discuss growth plans, motivation, way of working, and culture fit. Optional lunch with the team if visiting the office.

In the short term (1-3 years): grow into a Senior position, owning bigger parts of data science work; designing and executing complex analyses; scoping and driving medium- and long-term projects; contributing to strategic vision; working with team leads and product managers on planning.
In the longer term (3-5 years): choose between Manager track (leading teams, managing direct reports, contributing to company strategy, hiring) or Individual Contributor track (leading clinical and data science strategy, responsible for large analytical and product decisions, delivering high-quality data science work).

Software written in Python and Javascript (Typescript), with Azure as cloud provider.
Data science work primarily in Python: process clinical data with Pandas, Polars, and (Py)Spark; train and evaluate models with scikit-learn, scikit-survival, LightGBM.
Interpretability and explainability are central to model development.
Product stack includes FastAPI, SQL, VueJS, Docker.
No expectation to know all tools; opportunity to learn on the job.
Tech Funnel process allows team members to propose and evaluate new technologies for adoption.

Pacmed is hiring a Data Scientist (ML & Statistical Modelling)