Senior Robotic Learning Engineer (m/f/d)

The AI Teams at Agile Robots are looking for a Senior Robotic Learning Engineer (m/f/d), who will architect the full pipeline that turns raw robot and human data into deployed foundation models, from collection through post-training to on-robot deployment.

Your Responsibilities

• System Architecture: Own how collection, generation, augmentation, annotation, valuation, and evaluation fit together as one coherent data-to-model pipeline, and prioritize engineering investment where it most improves model performance.
• Model Evaluation & Deployment: Integrate post-training and evaluation frameworks, in simulation and on real hardware, and own the path to reliable, automatic deployment on robotic platforms.
• Data Valuation & Curation: Develop methods and metrics to assess data quality and value — diversity, coverage, redundancy, task relevance — and use them to guide what gets collected, kept, or discarded.
• Data Generation: Build synthetic and simulation-based data generation pipelines, including real-to-sim-to-real transfer, procedural task and scene generation, and generative models, to scale training data beyond physical collection alone.
• Data Collection: Design scalable systems for collecting both robot-generated data (teleoperation, autonomous rollouts, scripted policies) and human data (egocentric video, demonstrations) to train imitation learning and VLA models.
• Cross-Team Collaboration: Partner with research scientists, robotics engineers, and infrastructure teams to translate model and research needs into production data and evaluation systems.

Essential Skills

• Background: Master's degree or PhD in Computer Science, Robotics, Electrical Engineering, Machine Learning, or a related field, or an equivalent combination of formal training and professional experience.
• Professional Experience: 5+ years in robotics, autonomous driving, machine learning, or data engineering, with demonstrated ownership of systems spanning multiple pipeline stages, not a single component.
• Robot Learning: Hands-on experience with imitation learning, Vision-Language-Action models, robot foundation models, reinforcement learning, or other learning-based approaches to robotics.
• Programming: Software engineering proficiency in Python sufficient to build production-grade, reliable systems rather than research prototypes; familiarity with C++ is a plus.
• Systems-Level Thinking: Ability to reason about how a decision at one pipeline stage — what data is collected — affects outcomes at another, such as model performance or evaluation results, and to design across those boundaries.

Beneficial Skills

• Simulation: Experience with simulation platforms such as Isaac Sim or MuJoCo for synthetic data generation and sim-to-real transfer.
• Data Valuation Techniques: Familiarity with active learning, coreset selection, influence functions, or other methods for quantifying training data value.
• Human Data Collection: Experience with egocentric data collection, AR/VR devices, hand tracking, or teleoperation frameworks for embodied AI.
• Data Systems: Experience designing and operating large-scale data pipelines — collection, storage, processing, labeling — for machine learning training.
• Deployment & MLOps: Experience with model serving, edge or real-time inference, or cloud-based deployment infrastructure.