Staff AI Engineer - Robot Learning (Navigation)

Job description

Here at Humanoid, we believe in a future where robots amplify human potential. That’s why we’ve set out on a mission to build the world’s most capable, commercially-scalable, and safe humanoid robots. We’re bringing that mission to life with HMND‑01 Alpha - our rapidly developed humanoid platform now running in real industrial pilots - and we’re growing the team to take it even further. # About the Role We're hiring a **Staff AI Engineer, Robot Learning (Navigation)** to join our Perception and Navigation team based in London. In this role you will lead the design, development, and optimisation of next-gen robot learning systems for humanoid navigation, behavior learning, multi-agent interaction, and semantic goal reasoning in dynamic environments. We are interested in candidates who have a track record of driving end-to-end learned behaviours into production (e.g. self-driving, drones, robot navigation and other autonomous systems). At the Staff level, you aren’t just implementing existing models; you are defining the paradigm for how humanoids interact with a dynamic, unpredictable world. You will own the stack that transitions our robots from structured laboratory tasks to fluid, real-world autonomy. # What You'll Do - Develop next-generation learned navigation systems that integrate complex spatial reasoning and semantic goals to drive robust, real-world robot behaviors. - Work on open-ended navigation powered by Vision-Language-Action (VLA) models, enabling robots to understand context, navigate multi-agent environments, predict intent, and act safely in dynamic spaces. - Design and scale data pipelines and evaluation frameworks optimized for training large-scale, end-to-end (e2e) learned behaviors and multimodal navigation models. - Architect and deploy highly reliable ML systems, taking models out of simulation/labs and hardening them for predictable, repeatable execution on physical hardware. - Collaborate with cross-functional research and engineering teams to productionize large vision-language-action models, ensuring production metrics meet strict real-world reliability standards. - Stay ahead of the field, rapidly evaluate new model architectures, multi-agent strategies, and datasets to guide our embodied AI and behavior-learning roadmap. # What We're Looking For - Extensive experience in machine learning for embodied AI, with a proven track record explicitly focused on end-to-end (e2e) learned behaviors using large models (VLAs, VLMs, transformers, or diffusion). - Deep production expertise: You are someone who gets things into production that work reliably. You have hands-on experience deploying, monitoring, and optimizing large-scale ML systems. - Strong background in spatial reasoning and semantic goals, with experience handling multi-agent dynamics, crowding, or interactive environments. - Proficiency in PyTorch and the modern tooling required to train, fine-tune, and deploy large-scale foundation models for robotics. - Exceptional experimental and engineering skills, capable of taking ambitious behavior-learning concepts from initial research to rock-solid deployment on physical robots. - Comfortable working in a fast-moving, research-driven environment with evolving models, data, and tools. # What We Offer - Competitive equity: stock options with meaningful upside as we scale. - 30+ paid days off, including 23 days of annual leave, all UK bank holidays, and additional company closure days (including Christmas–New Year shutdown). - Private healthcare, including virtual and in-person care. - Pension scheme with 8% total contribution (5% employee, 3% employer) on full earnings. - Free daily breakfast, catered lunch, and snacks in-office. - Work at the frontier - collaborate daily with world-class engineers, researchers, and product experts building the next generation of AI and humanoid robotics. - Real ownership - direct access to founding leadership, meaningful input on product direction, and the ability to drive key initiatives from day one.