Internship - Humanoid Motion Generation (Diffusion or Flow Matching)

About Flexion

At Flexion, we're building the intelligence layer powering the next generation of humanoid robots. Our mission is to accelerate the transition from fragile prototypes to real-world deployment of humanoids. We are founded by leading scientists in robot reinforcement learning (ex-Nvidia, ex-ETH Zürich) and backed by leading international VC firms. Within a few months, we’ve gone from our first line of code to deploying real humanoid capabilities with our partners.

The Role

We’re looking for a talented MS or PhD intern to join our team in Zürich and push the boundaries of how humanoid robots perceive and interact with the world.

You will work at the intersection of generative whole-body trajectory generation (using diffusion models, flow matching, and/or auto-regressive models) and 3D environment perception.

If you enjoy solving open-ended problems in a fast-moving environment, this is a unique opportunity to learn from our world-class team of engineers and scientists, and contribute to our foundational software development.

Required skills

• Currently pursuing a Master’s or PhD degree in Computer Science, Robotics, Machine Learning, or a related technical field.
• Experience in diffusion models/flow matching or learning-based trajectory generation with relevant project experience.
• Familiarity with generative modeling and its application to time series data.
• Excellent Python and PyTorch skills; comfortable writing clean, modular code for rapid experimentation.
• Strong, proven analytical thinking and problem-solving skills.
• Ability to navigate and implement state-of-the-art methods from recent literature.
• Preferably: autonomous research/project work in human motion generation (neural avatars), robotic imitation learning, or ego-centric action prediction.

Additionally, the following skills are a plus but not required:

• Experience in robotics.
• Experience with simulation environments.
• Experience with multi-modal generative models.
• Experience with finetuning foundation models, e.g., Gr00t or SmolVLA, to produce whole-body actions or kinematics motions.
• Competitive compensation
• A front-row seat at one of Europe’s most ambitious robotics companies
• An energetic, collaborative team with a bias for action