(Senior) AI Research Engineer (m/f/d) Generative Video for Robotics

The AI Research Division of Agile Robots is looking for a (Senior) AI Research Engineer (m/f/d) focused on generative video and multimodal sequence modelling for robotics. In this role, you will develop temporally coherent generative models that support simulation, synthetic data generation, and downstream robot learning.

• Video Modelling: Build and optimize generative video models for robotics use cases such as synthetic data generation, predictive sequence modelling, and learning from embodied interaction.
• Multimodal Learning: Develop models that combine video with text, proprioception, and other structured signals to improve temporal reasoning and downstream usefulness.
• Evaluation: Design benchmarks and experiments that assess temporal coherence, sequence quality, and the practical value of generated outputs for robotics learning workflows.
• Data Pipelines: Build scalable training pipelines for large-scale video and sequence datasets with attention to quality, performance, and reproducibility.
• Research Translation: Apply advances in generative modelling, sequence learning, and multimodal AI to improve robotics-focused internal systems and research workflows.
• Collaboration: Work closely with robotics, simulation, and research teams to connect model development with real system constraints and applied use cases.

• Generative Video: Hands-on experience building generative video or temporal sequence models, including conditional generation, multimodal fusion, and temporal consistency optimization.
• Model Architectures: Strong practical knowledge of modern generative and sequence-modelling approaches such as diffusion models, autoregressive transformers, VAEs, GANs, or DiT-style architectures.
• Temporal Reasoning: Deep understanding of temporally coherent video generation, long-horizon sequence behaviour, and evaluation methods for predictive quality and stability.
• ML Engineering: Strong Python and PyTorch skills, including implementation of training pipelines, large-scale experimentation, and performance-oriented model development.
• Experimentation: Experience designing, running, and interpreting benchmarks on large-scale video datasets with clear judgment around model quality and failure modes.

• Robotics Applications: Familiarity with robotics-adjacent use cases such as simulation, synthetic data generation, or embodied learning workflows.
• Robot Data Modalities: Exposure to robot-relevant signals such as proprioception, force, tactile input, or other structured non-visual data used in embodied systems.
• Deployment Awareness: Experience bringing generative or multimodal models into production or production-near environments with attention to inference efficiency and scalability.
• Research Output: Publications, patents, or applied research contributions in generative modelling, multimodal learning, robotics, or computer vision.

• Dynamic high-tech company combined with financial soundness and world class investors.
• Join an interdisciplinary, international team with 60+ different nationalities in a collaborative work environment.
• Lots of development opportunities in the context of our continued growth.
• Challenging tasks and impactful projects alongside experts that enable professional and personal growth.
• Corporate Benefits Program that covers health, mobility and learning with 100 € net per month.
• Modern office facilites with a rooftop terrace overlooking Munich, free drinks & fruits, and regular company events contribute to a good working environment.