Humanoid Robotics Engineer

CloudWalk is seeking a Humanoid Robotics Engineer to join their ambitious R&D program focused on humanoid robotics. This role combines cutting-edge machine learning, robotics, and software engineering to develop advanced locomotion and high-level behaviors for humanoid robots. Based in São Paulo, Brazil, you'll work in a state-of-the-art robotics lab equipped with humanoid robots and vending machines.

The position requires expertise in reinforcement learning, simulation environments (NVIDIA IsaacLab/Sim/Gym), and real-world robotics deployment. You'll be responsible for the full development cycle, from designing simulation environments to implementing and deploying solutions on physical robots. The role demands strong technical skills in PyTorch, ROS 2, and the NVIDIA robotics ecosystem.

This is an exciting opportunity for someone passionate about embodied AI and autonomous systems. You'll work on challenging problems in robotics, from basic locomotion to complex obstacle navigation, using cutting-edge approaches like reinforcement learning, imitation learning, and transformer-based methods. The hybrid work model allows for both remote collaboration and hands-on lab work with physical robots.

The ideal candidate should have a strong foundation in machine learning, robotics, and software engineering best practices. Experience with legged locomotion, MuJoCo-based RL workflows, or advanced policy architectures would be advantageous. CloudWalk values diversity and inclusion, promoting a welcoming environment where all employees can be authentic regardless of their background.

This role offers the unique opportunity to work at the intersection of AI and robotics, contributing to groundbreaking research while implementing practical solutions. You'll be part of a team pushing the boundaries of what's possible in humanoid robotics, with access to advanced hardware and the latest software tools in the field.