(Senior) Machine Learning Engineer, Revenue Management (m/f/d)

About this role

Machine Learning Engineering at SIXT turns models into systems that decide prices, forecast demand, and steer a fleet of hundreds of thousands of vehicles, millions of times a day. You build these systems with our Data Scientists, run them in production, and shape the ML platform while it is still being built. The numbers you move are real. Transform the way the world moves, because people expect better. Join Team Orange in Lisbon.

YOUR ROLE AT SIXT You build ML systems together with Data Scientists, from the first design session to the validated outcome in production. One team, one goal. You own design and architecture: how features are computed once and served consistently, how predictions flow out, how training, serving, and retraining fit together. You get models live safely: offline evaluation, shadow runs, champion/challenger, rollback. You run what you ship.

You make quality structural: you define what a healthy ML system looks like, encode it as automated checks and alerts, and root cause issues so they do not recur. You are the bridge to MLOps: you bring the team's needs to the platform, bring platform standards back into the team, and co-own the templates and tooling that make the next use case faster. You multiply the team You set the engineering bar, design how coding agents execute the ML lifecycle from data preparation to deployment, and grow production practice together with Data Scientists.

YOUR SKILLS MATTER Experience You have shipped and operated ML systems in production for 3+ years and know where they break: training/serving skew, leakage, drift, silent degradation. Software Engineering You are fluent in Python, SQL and one more language such as Go or Java: you read, review, and debug production code with confidence, whatever agent wrote it. ML Systems Design You design clean contracts between data, model, and consumers and reason about latency, cost, and failure.

ML Fundamentals You understand forecasting, regression, and optimization well enough to challenge a model design and to recognize a wrong model in production. Software Factory You define acceptance criteria, orchestrate coding agents for implementation, QA, and security, and step in where they fall short. You review outcomes, not keystrokes, and want to shape how a team works this way. Platform Containers, orchestration, CI/CD, and model registries are daily tools.