Staff Machine Learning Engineer, Payment Optimization — Hybrid, Munich, Germany

Overview

Set the architecture and technical direction for machine learning that improves payment acceptance. This staff role works across product, data and payment engineering, identifying valuable problems and establishing systems other teams can reuse. The scope includes authorization performance, routing decisions and retry strategies, with responsibility extending beyond individual model development.

Requirements

• Candidates need strong production ML architecture, senior Python engineering and experience influencing teams without direct authority.
• Deep classical machine learning, cloud infrastructure and card-payment knowledge are required, including issuer behavior, authorization and retry logic.
• Experience owning low-latency systems and their reliability is essential.

Responsibilities

• Shape the technical roadmap and make investment proposals grounded in payment data and business outcomes.
• Design shared capabilities for features, model serving and experimentation.
• Establish standards for monitoring, model governance and reproducibility so that payment models can operate reliably as transaction volume and product coverage grow.
• Lead experimentation on live traffic, choosing suitable statistical methods for the decision being tested.
• Structure ambiguous problems into achievable work across teams and remove recurring architectural obstacles.
• Mentor senior engineers through design reviews and difficult implementation decisions, while improving documentation and technical collaboration across the machine-learning discipline.

Conditions

• This full-time role is hybrid in Munich, with three office days weekly.

Tech stack

• Python
• XGBoost
• LightGBM
• AWS or GCP
• MLOps