Machine Learning Research Engineer – Humanoid Robotics in City of London

Machine Learning Research Engineer

Location: London

Salary: Competitive, dependent on experience

Employment: Permanent

What if the reinforcement learning models you developed actually controlled a humanoid robot?

We're working with an innovative robotics company in London that's developing advanced humanoid robotic systems, and we're looking for a Machine Learning Research Engineer to help push the boundaries of what these machines can do.

This is an opportunity to work at the intersection of reinforcement learning, robotics, simulation and real-world control, helping develop intelligent behaviours that can be transferred from simulation onto physical robotic hardware.

If you've spent time developing learning-based control systems, training robotic policies or researching how robots interact with the physical world, this could be a particularly interesting next step.

The opportunity

You’ll be involved in researching, developing and implementing machine learning approaches that enable increasingly capable robotic movement and interaction.

Working alongside robotics, controls and software engineers, you’ll help bridge the gap between algorithms developed in simulation and behaviours deployed on real robotic systems.

What you’ll be doing

• Developing and training reinforcement learning policies for robotic control and movement.
• Building and improving simulation environments for robotic learning.
• Working on simulation-to-real transfer, including domain randomisation and policy robustness.
• Developing approaches to motion control, manipulation and potentially whole-body robotic behaviours.
• Evaluating and improving learning algorithms through experimentation and testing.
• Deploying and validating trained policies on physical robotic platforms.
• Collaborating with multidisciplinary teams across machine learning, robotics and control systems.
• Contributing to research that advances the capabilities of humanoid robots.

What we’re looking for

• Reinforcement learning, particularly for robotics or continuous control.
• Python and machine learning frameworks such as PyTorch.
• Robotics simulation environments such as NVIDIA Isaac Sim, Isaac Lab, MuJoCo or similar.
• Policy optimisation methods such as PPO, SAC or related algorithms.
• Simulation-to-real transfer, domain randomisation or system identification.
• Robot motion planning, kinematics, dynamics or control.
• Experience deploying algorithms onto physical robotic hardware.

A background in humanoid robotics, legged robotics, dexterous manipulation or embodied AI would be particularly relevant.

Who should apply?

You could be working as a Machine Learning Engineer, Robotics Research Engineer, Reinforcement Learning Engineer or Robotics Software Engineer.

We're also interested in speaking with MSc and PhD researchers who have practical experience applying machine learning to robotic systems and are considering a move into industry.

You don't necessarily need previous commercial humanoid robotics experience. Strong transferable experience in reinforcement learning, simulation and robotic control could be equally valuable.

Why consider this opportunity?

• Work on advanced humanoid robotics technology.
• Develop algorithms intended for deployment on physical robotic systems.
• Tackle challenging problems in reinforcement learning, motion and control.
• Collaborate with specialists across robotics, machine learning and engineering.
• Contribute to a rapidly developing area of intelligent robotics.

#J-18808-Ljbffr