Workster is partnering with a global mobility-tech leader to find a talented Data Scientist (Reinforcement Learning) to join their cutting-edge tech hub in Lisbon. In this role, you will design and implement advanced statistical models and real-time decision engines that directly power millions of dynamic pricing decisions per day.
You’ll work at the intersection of pricing strategy, statistical modeling, and scalable machine learning systems—owning the full lifecycle from analytical ideation to production deployment.
Your Role- Design advanced statistical models: Prototype regression-based pricing models (linear, GLM, Gaussian Process, mixed-effects) using a Bayesian framework
- Apply scalable inference methods: Use SVI, MCMC, and other techniques to process large-scale, high-volume, or streaming datasets
- Extend bandit algorithms: Improve multi-armed bandits with contextual features, richer priors, and uncertainty quantification
- Develop clean, reproducible pipelines: Build end-to-end pipelines for feature engineering, label generation, and automated data quality checks using Airflow or Dagster
- Package and deploy models: Use tools like FastAPI, Docker, or Kubeflow to serve modular ML services in production
- Monitor and detect anomalies: Set up performance dashboards and Bayesian control charts to catch data drift, overfitting, and anomalies in real-time
- Experiment and evaluate: Design A/B tests, multivariate experiments, or apply causal inference when randomization isn’t feasible
- Drive business impact: Collaborate with product managers and analysts to turn complex questions into measurable insights
- Share knowledge: Mentor peers, publish internal technical insights, and lead hands-on workshops on Bayesian workflows
- Strong statistical foundation: 5+ years applying regression, hierarchical models, or state-space methods in real-world settings
- Bayesian modeling expertise: Hands-on with PyMC, Stan, NumPyro, TFP, or similar libraries; confident building custom priors and likelihoods
- Experience in variational inference: Familiar with SVI, black-box VI, or advanced MCMC techniques at scale
- Software engineering mindset: Solid Python skills with familiarity in type hints, testing, and CI/CD best practices
- Cloud and orchestration fluency: Experience with AWS/GCP/Azure and tools like Docker, Kubernetes, or workflow schedulers
- Business communication skills: Capable of explaining uncertainty, lift, and risk clearly to both technical and executive audiences
- Continuous learner: You actively follow the latest developments in probabilistic ML and enjoy sharing knowledge with others
- Generous time off: 28 vacation days per year, your birthday off, and one volunteer day
- Work-life balance: Hybrid work setup, flexible hours, and no dress code
- Health & wellness: Private health insurance to support your well-being
- Perks & discounts: Coverflex benefits platform and discounts on transport, travel, and more
- Learning & development: Access to tech talks, external conferences, and training tailored to your growth