Machine Learning Engineer (m/f/d)

In this role, you will take a central position in building production-ready AI systems within an innovation-driven environment.

Your focus is on developing and validating new machine learning initiatives end to end, from early prototypes to adoption-ready solutions. You will design and implement scalable systems that process complex data streams across text, speech, and image, while ensuring they meet real-world requirements in terms of performance, reliability, and deployment constraints.

Working at the intersection of machine learning, data engineering, and product, you will build end-to-end solutions that connect data ingestion, model inference, and user-facing applications. Your work enables teams to bring new AI capabilities into production, including in highly constrained environments such as on-premise and air-gapped systems.

- Build end-to-end ML prototype systems from data ingestion to inference and product integration (e.g. search, insights, UX flows) - Develop and deploy machine learning models in production-like environments - Design and implement scalable inference pipelines (e.g. Triton or TorchServe) - Create evaluation frameworks including datasets, baselines, and metrics such as quality, latency, and cost - Provide clear recommendations based on evaluation results - Package solutions for adoption through containerization, reproducible deployments, and clear documentation - Implement RAG pipelines using vector databases such as pgvector or Milvus - Build streaming and real-time analytics pipelines for use cases like video or speech processing - Optimize models for performance using techniques such as quantization, pruning, or distillation - Contribute to MLOps processes including monitoring, A/B testing, rollbacks, and lifecycle management - Ensure operational readiness through proper instrumentation, monitoring, and handling of sensitive data - Collaborate closely with Data Scientists, Research, Data Engineering, and DevOps teams

- 3 or more years of experience in Machine Learning Engineering or a similar role - Strong Python skills and hands-on experience with PyTorch and/or TensorFlow - Proven experience deploying ML models into production (e.g. NLP, ASR, Computer Vision) - Solid experience with Docker and Kubernetes in production-like environments - Experience building API-based ML services and CI/CD pipelines - Strong understanding of performance trade-offs across latency, cost, and scalability - Experience with evaluation, monitoring, and experiment tracking - Ability to move models from experimentation to stable, reproducible systems - Structured and pragmatic working style with strong ownership - Eligibility to work in Germany and based in Germany

- Experience with LLMs and RAG systems - GPU serving and optimization (e.g. Triton, ONNX, TensorRT, CUDA) - Experience with streaming or data pipeline tools such as Kafka, Ray, Spark, Flink, or Beam - Familiarity with MLflow or similar MLOps tools - Experience in regulated or public-sector environments - German language skills (B1 or higher)

What You Can Expect

- Work on production-grade AI systems with direct impact - A modern ML stack including Kubernetes, streaming systems, and hybrid or on-premise deployments - Exposure to challenging environments such as air-gapped or high-security setups - High ownership and visibility across multiple teams - Access to modern GPU infrastructure and tooling - Opportunity to shape ML and MLOps standards from the ground up - Remote-first setup within Germany with regular in-person sessions in Berlin - 30 days of vacation plus equipment and learning budget

You are interested? Feel free to get in touch directly: Constantin Clodius Phone: +49 170 3660 753 E-Mail: E-Mail: c.clodius@zabelglobal.com c.clodius@zabelglobal.com

4 November 2025

Technology - SaaS (Software as a Service)

€ 130.000 p.a.

+49 170 3660753

https://www.linkedin.com/in/constantin-clodius/