ML Engineer – Reinforcement Learning & Multi‑Armed Bandits (Hybrid, Lisbon)

About the role

We are looking for a senior Machine Learning Engineer to join our global mobility‑tech leader in Lisbon. You will focus on Reinforcement Learning and Multi‑Armed Bandit algorithms, building systems that drive millions of real‑time decisions.

Key responsibilities

• Design, prototype and optimise contextual multi‑armed bandit models with uncertainty quantification.
• Implement, scale and productionise reinforcement‑learning pipelines for pricing, recommendation or allocation.
• Drive the full ML lifecycle from feature engineering to deployment and monitoring.
• Apply scalable inference techniques such as SVI, MCMC and variational inference on large datasets.
• Develop reproducible Python workflows using PyMC, TensorFlow Probability or NumPyro.
• Integrate models into cloud‑native, containerised environments (Docker, Kubernetes, Airflow/Dagster).
• Monitor model performance, detect data drift and anomalies with Bayesian control charts and dashboards.
• Collaborate with product, analytics and engineering teams to translate business needs into ML solutions.
• Mentor peers, run workshops and contribute to technical documentation.

Required profile

• 5+ years of hands‑on experience delivering statistical or ML models in production.
• Proven expertise with reinforcement learning or online decision‑making systems.
• Strong background in probabilistic modelling and Bayesian inference.
• Fluent English and eligible to work in Portugal, with regular on‑site attendance.

Required skills

• Python
• PyMC, Stan, NumPyro, TensorFlow Probability
• SVI, MCMC, black‑box VI
• AWS, GCP or Azure
• Docker, Kubernetes
• Airflow or Dagster
• CI/CD and automated testing

What we offer

• Competitive compensation package.
• 28 days of holiday.
• Hybrid work model with a vibrant, diverse team.
• Opportunities for continuous learning and research publication.