Senior Machine Learning & Data Engineer (m/w/d)
MLOps/ML engineer at Airbus Defence and Space's new Backnang (Stuttgart area) center, building pipelines and the MLOps platform to train, evaluate, version, and deploy ML models for autonomous UAV systems. Core stack: Python, PyTorch, ONNX, Kubernetes, GitLab, MLflow, Kubeflow in multi-cloud and air-gapped clusters.
- Compensation
- €70K–€100K / yr
- Location
- Mitte, Stuttgart
- Work setup
- Not specified
- Seniority
- Senior
- Employment
- Not specified
- Experience
- Not specified
- Category
- Data engineering
- Posted
Skills
Must have
- Completed degree in a STEM field (CS, math, electrical engineering, aerospace, or AI)
- Professional experience in MLOps/ML engineering with models and platforms used in production
- Solid ML understanding: model training, evaluation, improvement; hands-on experience with PyTorch and formats like ONNX
- Very good Python skills and commitment to clean, maintainable software
- Experience with Kubernetes, CI/CD, and MLOps tools such as GitLab, Kubeflow, MLflow, KServe, and artifact registries
- Systems thinking: keeping sight of the overall solution across models from many sources
- Team spirit, clear communication, and interest in shaping an emerging platform
- Fluent (business-fluent) German and English
- Awareness of compliance risks and integrity in behavior
Nice to have
- C++ knowledge
- Model optimization for edge inference (e.g., TensorRT, quantization)
- Experience with hardware-in-the-loop environments
- Simulation environments and synthetic data
- Distributed training on GPU clusters
Full description
Salary: 70.000 - 100.000 € per year Requirements: Abgeschlossenes Studium im MINT-Bereich, z. B. Informatik, Mathematik, Elektrotechnik, Luft- und Raumfahrttechnik oder Künstliche Intelligenz Berufserfahrung in MLOps oder ML Engineering mit Modellen und Plattformen, die aktiv für produktive Systeme genutzt werden Fundiertes ML-Verständnis: Kenntnisse darüber, wie Modelle trainiert, evaluiert und verbessert werden, sowie praktische Erfahrung mit PyTorch und gängigen Formaten wie ONNX Sehr gute P…