Machine Learning Engineer – Whole-Body Control

About the role

The Bot Company is building high‑performance whole‑body controllers that enable robust, agile motion and manipulation for home robots. As a Machine Learning Engineer you will design, train, and deploy low‑level control policies, owning the full pipeline from simulation environment creation to large‑scale training and sim‑to‑real transfer.

Key responsibilities

• Train whole‑body policies for locomotion, manipulation, and coordinated motion.
• Build scalable simulation environments using IsaacLab, MuJoCo, or similar frameworks with parallel rollouts.
• Design reward functions and curricula to support stable long‑horizon learning.
• Manage sim‑to‑real transfer through domain randomization and structured evaluation.
• Run, debug, and optimize large‑scale GPU training experiments.

Required profile

• Very strong coding skills in Python, C++, or Rust (CUDA a plus).
• Solid foundation in modern reinforcement learning.
• Experience training embodied control policies and hands‑on robotics simulation.
• Strong intuition for dynamics, contact, and actuation.
• Comfortable running large‑scale GPU experiments.

Required skills

• Python
• C++
• Rust
• CUDA
• Reinforcement learning
• Robotics simulation (IsaacLab, MuJoCo)
• GPU training
• Dynamics and contact modeling
• Actuation control

What we offer

• Work with a small, elite team focused on rapid iteration and fast execution.
• Broad ownership and direct responsibility across the stack.
• Opportunity to solve challenging robotics problems at scale.