Staff Data Scientist Reinforcement Learning – Offer Intelligence (m/f/d)

Description

The team sits where pricing science, offer personalisation, advanced statistics and scalable software meet. You will build probabilistic models, such as hierarchical Bayesian regression, SVI or bandit algorithms, and move them into production, where they set prices and offers for millions of decisions each day.

Beyond pricing, the team is building the layer that decides which offer and which experience each customer sees, so your models will influence both what is charged and what is shown. You own the work from framing the analytical question to running a live model.

The company is a global mobility provider with revenue of €4.3 billion and about 9,000 employees. Its platform ONE brings together car rental, car sharing, taxi, ride and chauffeur services, and car subscriptions, with a fleet of 360,000 vehicles, 4,000 cooperation partners and around 5 million drivers. With franchise partners, it operates in more than 100 countries across 2,000 rental stations.

Candidates who do not meet every point in the list below are still welcome to apply.

Responsibilities

• Prototype and build regression models, including linear, GLM, mixed-effects and Gaussian Process, using a Bayesian approach that guides pricing strategy at scale.
• Design personalisation and recommendation systems, such as two-tower retrieval, embedding-based candidate generation, learning-to-rank and sequential or transformer models, and put them into production.
• Use contextual bandits and off-policy evaluation so these systems keep learning from real user interactions.
• Apply SVI, MCMC and importance sampling to make sound decisions under uncertainty, including with sparse, high-volume or streaming data.
• Build and maintain reproducible pipelines for feature engineering, label generation and automated data validation with Airflow or Dagster, keeping code quality and reliability high.
• Deploy models with FastAPI, Docker or Kubeflow, and set up real-time monitoring with Bayesian control charts and dashboards to spot drift and anomalies early.
• Design A/B and multivariate tests, apply quasi-experimental and causal impact methods, and measure the value of information to balance exploration and exploitation.
• Work with product, revenue and customer experience leads to turn loose ideas into testable hypotheses, and mentor colleagues while raising the team's Bayesian practice through internal publications and workshops.

Requirements

• Deep statistical skill, with 5+ years of applying regression, hierarchical or state-space methods in production.
• Working knowledge of Bayesian tools such as PyMC, Stan, NumPyro or TFP, and comfort writing custom likelihoods and priors.
• Proven experience scaling inference with SVI, black-box VI or advanced MCMC on large datasets.
• Nice to have: experience with recommender and personalisation stacks, including two-tower or embedding retrieval, learning-to-rank, sequential models and contextual bandits, plus counterfactual or off-policy evaluation that proves incremental value rather than chasing click-through.
• Production-grade Python, with type hints, testing frameworks and CI/CD workflows.
• Hands-on work with AWS, GCP or Azure, plus containerisation (Docker or Kubernetes) and workflow orchestration.
• Ability to explain uncertainty, lift and risk in plain English to executives, while working in depth with scientists and engineers.

Conditions

Office days: two per week, with flexible hours.

Location: Lisbon, at Factory Lisboa in Beato or ARTs Lisboa in Parque das Nações.

Equal-opportunity statement: applicants are considered regardless of background.

Benefits

• Events with internal and external speakers, supporting continuous learning and open exchange across teams and borders.
• 28 vacation days per year, an extra day off on your birthday, and one volunteer day each year.
• Private health insurance that also covers your spouse and children at no cost.
• Yearly Coverflex allowance and a daily meal allowance.
• Exclusive rates on all the company services, with special discounts for family and friends.