Datascientist

Perifit is a profitable and fast-growing femtech startup based in Paris, France. Turnover was multiplied by 100 in 5 years!

Our goal is to assist women worldwide with some of the most challenging and under-addressed moments in their lives, using fun, effective and science-based self-care solutions.

Our first product, Perifit Care — a connected pelvic floor trainer and its mobile games — launched in 2018 and has become the #1 connected solution for pelvic floor training. It has helped hundreds of thousands of women improve their pelvic health and is the best-ranked app in its category.

We now develop innovative pelvic floor trainers and breast pumps using advanced hardware, software and data technologies to help women improve their lives. Our mission is to empower women to take control of their health and improve their overall well-being.

Our team grew from 2 co-founders to 50+ employees, and we operate in a truly global environment, with customers and partners around the world.

At Perifit, we stand for equality, inclusivity, benevolence and wellbeing for both our customers and our employees.

Joining Perifit means being part of a mission-driven team that is making a real difference in women's lives, while contributing to one of the most ambitious femtech companies in Europe.

We’re very proud of our values

FREEDOM – FEMINISM – STARTUP MINDSET – REAL IMPACT – NO EGO

Why we need you

Perifit is entering a new phase: we want to use our unique combination of connected devices, real-world pelvic floor biomechanical data, app usage data, symptom questionnaires and long-term progression data to improve the medical effectiveness of our products.

Our ambition is to better understand what helps each woman progress, stay engaged and improve her symptoms, then turn that understanding into smarter models, better product decisions and more personalized training journeys.

In 2026, we want to build the data and applied ML foundations that will help us:

• better understand our users, their behavior and their progression
• improve retention and training adherence
• improve the quality of our sensor models using real-world data
• identify which signals are truly useful to improve medical effectiveness
• build the foundations for future AI-powered coaching and personalization

We are looking for a senior, pragmatic and hands‑on applied ML leader to own this strategy. You will help us decide what to build, what to measure, what to deploy, and just as importantly, what not to do.

What you will be doing

You will lead the data and applied ML work behind Perifit’s AI strategy. Your scope will cover data foundations, churn and retention models, sensor model improvement and the first building blocks of future AI-powered personalization. You will turn ambiguous business and product questions into clear datasets, metrics, models and production‑ready systems.

Concretely, you will define the right data events, cohorts and labels; build pragmatic models that can improve retention and adherence; help improve our sensor models with real-world usage data; and make sure models can be deployed, monitored and maintained in production, including through APIs, backend services or local inference in the mobile app when relevant.

You will also coordinate the internal test benches and physical jigs that generate the data used to train and validate our sensor models. You will not be expected to be a hardware engineer, but you should be comfortable working at the intersection of Python, embedded software, sensors and physical test setups, and helping us make these benches more reliable, repeatable and scalable internally.

This is a highly cross‑functional role. You will work closely with:

• Product, to translate data insights into user‑facing features
• Software, to integrate models into production systems
• Research, to connect ML work with biomechanical and symptom data
• OBS and Data Engineering, to structure the datasets, pipelines and operational insights needed to understand users and improve retention

What hard skills we need to see

Educational background: Engineering, computer science, applied mathematics, statistics, data science or equivalent experience.

Relevant experience: 7+ years of experience in data science, applied ML, ML engineering or data‑heavy product engineering.

Python and SQL: Strong hands‑on experience.

Data analysis: Strong ability to explore messy datasets, build cohorts, define labels, identify bias, produce clear visualizations and communicate what the data does and does not say.

Data engineering basics: Experience working with modern data warehouses and production datasets. BigQuery is a plus. dbt, Metabase or equivalent tools are a plus.

Applied ML: Solid experience with classical ML models, feature engineering, model evaluation, model monitoring and production constraints.

Test benches and sensor data generation: Comfortable working with physical test setups, internal jigs or benches that generate the data used to train and validate models. You do not need to be a hardware engineer, but you should be able to coordinate improvements with embedded software, hardware and engineering teams.

Production deployment: Ability to put a model into production yourself, at least through a simple backend, API, serverless deployment, batch pipeline or equivalent.

Product judgment: Ability to transform a vague ambition into a clear problem statement, a dataset, a baseline model, a metric and a decision framework.

Large‑scale usage: Experience working on products, models or data systems used by a large number of users. Ideally hundreds of thousands of users or more.

What kind of person we are looking for

Mindset: Pragmatic, structured and intellectually honest. You care more about solving the right problem than using the most impressive model.

You know when not to use ML: You are comfortable saying “this is not a ML problem yet”, “the data is not good enough”, “the label is too weak”, or “a rule‑based system is better for now”.

You are pedagogical: You can explain complex technical trade‑offs to non‑technical leaders. You know how to align expectations without killing ambition.

You are end‑to‑end: You are not only a notebook person. You can go from raw data to analysis, from analysis to model, from model to deployment, and from deployment to monitoring.

You are product‑minded: You care about user impact, not just model metrics. You understand that a churn model is only useful if it leads to actions that improve retention.

You are comfortable with ambiguity: You can take a vague idea like “AI should personalize the training journey” and turn it into a clear set of hypotheses, experiments, constraints and next steps.

You are a feminist: You believe women deserve better health solutions. You are comfortable working on topics such as pelvic floor health, childbirth, urinary leaks, female intimacy and women’s health in general.

Language proficiency: Fluent in English, with strong written communication skills. French is a plus but not mandatory.

You will shine if

• You have experience with IoT, connected devices, sensors or hardware‑generated data.
• You have worked on consumer apps, subscriptions, churn, retention or lifecycle personalization.
• You have deployed models to large user bases and understand production constraints, monitoring, rollback, model drift and inference cost.
• You have worked with mobile inference, embedded inference, ONNX, Core ML, TensorFlow Lite or similar tools.
• You have experience building data products from scratch in a startup or scale‑up environment.
• You have experience with A/B testing, causal inference, quasi‑causal analysis or experimentation frameworks.
• You have worked in healthtech, digital health, medtech or another domain where model quality and user trust matter.
• You have coached junior data scientists, analysts or engineers without necessarily being a people manager.

Why this is an offer you can’t refuse

This is a rare opportunity to build the AI and data foundations of a product already used by hundreds of thousands of women.

You will not join a company trying to invent an AI strategy out of thin air. Perifit already has the ingredients of a strong AI moat: connected hardware, real-world biomechanical data, longitudinal usage data, symptom outcomes, a large installed base and a product where personalization can genuinely improve user experience.

Your role will be to turn this potential into a real, defensible and useful system.

You will have high ownership, direct access to the CTO, and the freedom to shape the technical roadmap. You will be expected to challenge ideas, simplify overambitious plans and build systems that work in the real world.

If you want to work on applied ML with real product impact, real users, real‑world sensor data and a mission that matters, this role is for you.

Advantages

• Comprehensive Health Coverage: Excellent mutual health insurance — Alan — 100% financed by the company.
• Meal Benefits: Swile “restaurant tickets”, 60% covered by the company.
• Wellness Support: Free subscription to ClassPass to stay healthy and energized.
• Flexible and Friendly Work Environment: Spacious offices in the heart of Paris with a welcoming atmosphere.

Keywords

Head of Data, Applied ML, Lead Applied ML Engineer, Senior Data Scientist, ML Engineering, Data Science, BigQuery, Python, SQL, Churn Prediction, Retention, Sensor Data, IoT, Femtech, Healthtech, ONNX, Model Deployment.