Senior AI Infrastructure Engineer (Zürich, 100%)

Loki Robotics seeks a Senior AI Infrastructure Engineer in Zürich to design and maintain the ML systems supporting robot learning, including data pipelines, training infrastructure, and model deployment.

This position focuses on constructing the foundational machine learning infrastructure that enables robot learning. The engineer will span the entire lifecycle, from ingesting raw data and building internal tools to deploying models and ensuring their safe operation in production environments.

Responsibilities

• Develop a multi-layered data processing system that transforms raw inputs into versioned datasets for training.
• Establish and manage training infrastructure using cloud-based GPU resources and containerized environments.
• Deploy models to both cloud and edge environments while creating evaluation tools to ensure deployment safety.
• Design and upkeep internal tools for monitoring data quality and performing inspections.

Requirements

• Over five years of professional software engineering experience, with a strong emphasis on data platforms or ML infrastructure.
• Proficiency in Python and comfort working with typed languages such as Rust, Go, or C++.
• Solid background in data pipelines and storage solutions, including tiered architectures, workflow orchestration, schema evolution, and backfills.
• Experience provisioning GPU instances in the cloud and training models in reproducible, containerized setups with a model registry.
• Practical ML experience, including understanding dataloader throughput, GPU utilization, and debugging stalled or slow training runs.
• Strong software engineering practices, including infrastructure as code, code reviews, architectural refactoring, and building automation tools.

Nice to have

• Experience with edge inference on devices like Jetson, using TensorRT, ONNX, or quantization techniques.
• Familiarity with multimodal and time-series data, such as video pipelines, sensor logs, MCAP formats, and time alignment.
• Knowledge of distributed training and performance optimization strategies.
• Skills in managing GPU clusters and orchestrating jobs.
• Production experience with Rust.

About the company

Loki Robotics fosters a culture that prioritizes rapid learning and a deep commitment to building high-quality products. The team values individuals who are eager to grow and dedicated to excellence in their work.

• Emphasis on fast learning and continuous improvement.
• Focus on building great, high-quality products.