Senior AI Engineer Self-Supervised Learning

10 months ago

Zürich, SwitzerlandSenior

Responsibilities

Design, build, and maintain scalable pipelines for processing, filtering, and transforming terabytes of multimodal sensor data from autonomous robots.
Develop self-supervised and representation learning algorithms to extract features, discover patterns, and generate pseudo-labels from unlabeled data.
Collaborate with VLA Foundation Model and Reinforcement Learning teams on data requirements, APIs, and dataset strategies.
Design evaluation strategies, benchmarks, and datasets for measuring data pipeline quality and downstream model performance.
Own data integration workflows, efficient DataLoaders, and data access patterns for model training and experimentation.
Research and prototype data curation, active learning, anomaly detection, and related techniques.

Master’s degree or higher in Computer Science, Machine Learning, Robotics, or a related field.
At least three years of industry or research experience, with PhD experience applicable.
Deep expertise in self-supervised and representation learning for multimodal sensor data, including techniques such as contrastive learning, masked autoencoders, and world models.
Experience building and managing large-scale machine learning data pipelines using tools such as Spark, Kubeflow, or similar cloud-native tools.
Strong understanding of robotic sensor data, including camera, LiDAR, IMU, and odometry.
Strong Python programming skills and deep PyTorch experience, including creating custom and efficient DataLoaders.
Experience with MLOps practices and data versioning tools such as DVC or Pachyderm.
Preferred qualifications include a PhD or equivalent research experience, publications at leading ML or robotics conferences, and experience with GANs or Diffusion Models for data augmentation or simulation.

51-200 employees