machine learning engineer

PDR.cloud is a Berlin-based SaaS company that provides cloud-based software for automotive repair service providers to manage billing and damage processes. Its platform combines classic dealer-management system functions with smart solutions and modern IT architecture to enable fully digital workshops.

• Own the complete ML lifecycle, including exploratory analysis, data modeling, feature engineering, model development, and production pipelines;
• Handle data acquisition, schema design, data preparation, and data evaluation tools;
• Collaborate with Product and Engineering to validate hypotheses, iterate on models, extract insights, and turn them into product decisions;
• Shape architecture, tooling, and data culture;
• Calibrate models for a growing, heterogeneous customer base;
• Continuously improve model quality.

• Several years of experience as a Machine Learning Engineer, Data Engineer, or Applied Data Scientist with a clear engineering focus;
• Ability to design pipelines independently and solve data problems autonomously;
• Strong Python and SQL skills;
• Confident use of AWS, GCP, or Azure;
• Solid statistical foundation covering classical and probabilistic model families;
• Understanding of when to apply different statistical approaches;
• Experience with production model deployment and MLOps fundamentals;
• Ability to build new model pipelines from scratch with operations professionals;
• Pragmatic, hands-on mindset and confidence working with incomplete data and shifting requirements;
• Fluent German;
• Fluent English;
• Nice to have: experience with event or sequence data and irregular timestamps, knowledge of probabilistic modeling or process mining, experience building data and ML infrastructure in a greenfield setup, background in SaaS, automotive, or insurance environments, experience in agile product teams and direct collaboration with pilot customers.

• Work authorization for Germany is required;
• Regular workshops and team-building events in Berlin;
• Flat hierarchies and short decision-making paths.