All3

About the Company

All3 is a construction technology company that’s redefining how buildings are designed and built. By combining AI-powered design with robotic fabrication, All3 enables developers to deliver custom, sustainable buildings at the cost and speed of mass production. At its core, All3 is about solving the construction industry’s biggest challenges - low productivity resulting in high costs, and environmental impact - by bringing the precision and productivity of advanced automation to the built environment. In essence, All3 isn’t just a construction company - it’s a technology-driven enabler that empowers architects to design freely, developers to build more efficiently, and communities to access high-quality, sustainable housing.

Listed Jobs

Company Name

All3

Job Title

Reinforcement Learning Control Engineer

Job Description

Job Title: Reinforcement Learning Control Engineer Role Summary: Design, train, and deploy reinforcement learning control agents for autonomous locomotion and manipulation on the Mantis legged robot, ensuring robust performance from simulation to real‑world operation. Expectations: Deliver reliable learning‑based control policies that integrate seamlessly with existing perception, state estimation, and software systems; provide continuous improvement of hardware performance through iterative testing and analysis. Key Responsibilities - Design and train RL policies using sensor models (LiDAR, cameras, IMUs) for locomotion and manipulation tasks. - Integrate learned components into a complex software stack while maintaining system stability and real‑time performance. - Develop and validate sim‑to‑real transfer strategies and deploy policies on physical hardware. - Analyze hardware experiments, identify failure modes, and iterate to enhance robustness and safety. - Collaborate with perception, state estimation, and software teams to deliver cohesive, real‑time robotic behavior. - Contribute to the evolution of a unified whole‑body control framework for locomotion‑manipulation integration. Required Skills - Proficiency in Python and PyTorch; solid C++ experience. - Deep understanding of ML techniques, challenges, and practical constraints in physical systems. - Hands‑on experience with legged locomotion (quadruped or biped) or robotic manipulation. - Knowledge of robot dynamics, kinematics, and contact modeling. - Demonstrated experience deploying learning‑based components on real robotic hardware. - Ability to reason about stability, safety, and robustness beyond reward optimization. Required Education & Certifications - Bachelor’s degree in Robotics, Computer Science, Electrical Engineering, or a related field (Master’s or PhD preferred). - No specific certifications required.

London, United kingdom

On site

27-02-2026