Staff MLOps Engineer (f/m/d)

You will join the team building the ML infrastructure behind MicroVision's automotive LiDAR perception software: models that have to run in real time, on embedded hardware, in safety-relevant systems. In this role, you will be the technical owner of the ML pipeline, setting its direction and pushing the existing solution to the next level.

Your focus:

• End-to-end training pipeline: design and build the on-prem ML pipeline covering the full training lifecycle
• Model optimization & edgedeployment: pruning, quantization, and deployment to edge devices for real-time LiDAR applications
• Model evaluation and benchmarking on our servers as well as on our target devices
• Self-service for ML engineers: build platform capabilities that let ML engineers deploy independently, working cross-functionally with ML, Data, and DevOps

Your profile:

• MLOps experience for edge devices: you built ML training pipelines and model registries that teams rely on to build products
• You understand throughput, latency tradeoffs, and GPU resource management
• 3+ years of experience in MLOps or ML pipeline engineering, including production ownership
• Ownership mindset: you don't just surface problems, you own the solution
• Platform engineering mindset: you care about developer experience, write documentation others use, and treat reliability and observability as first-class concerns
• Fluent English (German is a plus)

Tech stack you can expect:

• Languages & Frameworks: Python, C++, CUDA, TensorRT, PyTorch, Tensorflow