fb-pixel

Humanoid Robotics Engineer

CloudWalk

São Paulo
View Job

Description

Help us build CloudWalk's ambitious R&D program around humanoid robotics. You'll own the full cycle of developing locomotion and high-level behaviors: design tasks and environments in NVIDIA IsaacLab/Sim/Gym, train policies with RL/IL, validate on simulation and deploy on our physical humanoid robot (Sim2Real).

The Robotics Division

  • We work on embodied AI and autonomous systems across various robotics platforms and applications
  • Full development cycle from simulation environment design in NVIDIA IsaacLab/Sim/Gym to deployment on physical robots
  • São Paulo lab with advanced robotics hardware including humanoids and vending machines

    • What you'll do:

  • Invent and build humanoid tasks and environments – from simple flat-ground locomotion to challenging obstacle courses
  • Train and refine policies using RSL-RL, RL-Games, Stable-Baselines3, or similar GPU-parallel RL libraries
  • Explore cutting-edge approaches – reinforcement learning, imitation learning, diffusion policies, and transformer-based approaches for continuous control
  • Close the Sim2Real gap through domain/dynamics randomization, calibration, and staged deployment on the humanoid platform in our lab
  • Integrate with robotics ecosystem – ROS 2 (ros2_control, MoveIt 2) and NVIDIA's robotics stack (Isaac ROS, cuRobo/cuMotion, NVBlox)
  • Orchestrate complex behaviors with BehaviorTree.CPP/Groot, and experiment with Isaac GR00T for high-level humanoid skill composition
  • Measure and iterate with reproducible pipelines, dashboards, and design documentation to accelerate R&D cycles

    • What you'll need:

  • Solid ML and RL foundations (PPO, SAC, TD3) with hands-on PyTorch experience
  • NVIDIA Isaac ecosystem proficiency – Isaac Sim/Isaac Lab and legacy environments (Isaac Gym, OmniIsaacGymEnvs, Orbit)
  • Strong ROS 2 skills – ros2_control, lifecycle nodes, planning and perception integration
  • Sim2Real expertise – domain randomization, dynamics tuning, and safe hardware deployment
  • Engineering best practices – version control, containerization, reproducibility, and documentation
  • Must be able to work on-site in São Paulo for regular robotics lab sessions with physical robots

    • Bonus points if you have:

  • Legged locomotion experience with humanoid control tasks (Legged Gym, ETH Zurich RSL environments)
  • MuJoCo-based RL workflows (Gymnasium, dm_control, robosuite) for algorithm prototyping and benchmarking
  • Advanced policy architectures – Diffusion/Transformer policies for robotics applications
  • NVIDIA robotics stack – GPU-accelerated planning (cuRobo, cuMotion) and Isaac ROS perception (NVBlox, VSLAM)
  • Real-time systems and mechatronics knowledge (PREEMPT_RT, C++ optimization)
  • Strong portfolio of robotics and ML projects (GitHub repos, demonstrations, research papers)

    • Work Model

  • Hybrid role – collaborate remotely but spend regular days in our robotics lab
  • Hands-on experimentation with physical humanoid robots is a core part of this job
  • Test, validate, and deploy your simulation work on real hardware in our São Paulo lab

    • Recruiting Process

  • Challenge: You'll receive a technical challenge to demonstrate your skills in robotics, simulation, and machine learning
  • Technical interview
  • Cultural interview
    • Note: If you are not willing to take an online quiz and demonstrate your technical capabilities with both simulation and physical robotics systems, do not apply.
    Diversity and inclusion:
    We believe in social inclusion, respect, and appreciation of all people. We promote a welcoming work environment, where each CloudWalker can be authentic, regardless of gender, ethnicity, race, religion, sexuality, mobility, disability, or education.

    Technologies

    GitHubC++PyTorchGymOrchestrateAble

    Nice to Have

    Experience with legged locomotion and humanoid control tasks, especially using platforms like Legged Gym or ETH Zurich RSL environments.Knowledge of MuJoCo-based RL workflows for prototype development and benchmarking.Familiarity with advanced policy architectures, such as Diffusion/Transformer policies tailored for robotics applications.Experience in NVIDIA's robotics stack, specifically with GPU-accelerated planning and perception tools like cuRobo, cuMotion, and NVBlox.Understanding of real-time systems and mechatronics, specifically in areas such as PREEMPT_RT and C++ optimization.A strong portfolio demonstrating previous robotics and machine learning projects, including GitHub repositories, experimental demonstrations, or published research papers.

    Must Have

    Solid foundations in Machine Learning (ML) and Reinforcement Learning (RL) including algorithms like PPO, SAC, TD3, and hands-on experience with PyTorch.Proficiency in NVIDIA Isaac ecosystem including Isaac Sim and Isaac Lab, as well as familiarity with legacy environments like Isaac Gym and OmniIsaacGymEnvs.Strong skills in ROS 2, particularly in areas like ros2_control, lifecycle nodes, and integration for planning and perception.Expertise in Sim2Real techniques including domain randomization and dynamics tuning for safe hardware deployment.Understanding and implementation of engineering best practices such as version control, containerization, reproducibility, and thorough project documentation.Willingness and ability to work on-site in São Paulo for collaborative work in the robotics lab.
    Apply for this position