Robotics Learning Engineer – Humanoid Robotics / Deep Learning

Latent Edge Ltd; City Of London; On-site; GBP 90,000 - 130,000; Full time

Benefits offered by this job

Competitive salary

Equity participation

30+ days leave

Private healthcare

Pension contributions

Free daily breakfast and snacks

Direct hardware exposure

Technical ownership

Fast-moving environment

Job summary

Latent Edge Ltd. in London is seeking a Robotics Learning Engineer to advance humanoid robotic manipulation and deep learning capabilities on real hardware.

You will drive learning systems from training to deployment, enabling robots to interact with and grasp objects in real-world environments. Join a team at the forefront of humanoid robotics where ML, control and hardware converge to deliver industrial-grade capabilities.

Qualifications

• 3+ years working with robotic systems in industry or research.
• Hands-on experience deploying software on real robots.
• Experience with deep learning and neural network training or fine-tuning.
• Strong Python skills and familiarity with ML frameworks.
• Understanding of robotic control, teleoperation or motion planning.
• Experience handling ML datasets, training and evaluation.
• Strong software engineering skills and maintainable code.

Responsibilities

• Develop and fine-tune robot manipulation policies using behavior cloning and RL.
• Train, evaluate and deploy ML models on real robotic hardware.
• Improve training datasets through preprocessing and quality optimization.
• Collaborate with sim-to-real teams to improve environments.
• Work with data collection teams to improve data quality and coverage.
• Develop new observation and action spaces for advanced manipulation.
• Improve robot movement, responsiveness and smoothness with controls engineers.
• Investigate hardware limitations and contribute to future design improvements.
• Run experiments, analyze results and identify improvements.

Skills

Python

PyTorch

Machine learning

Robotics

Reinforcement learning

Control systems

Teleoperation

Experimentation

Data analysis

Software debugging

Tools

JAX

OpenVLA

Job description

Robotics Learning Engineer – Humanoid Robotics / Deep Learning

Location: London, UK

Salary: Competitive salary + equity + benefits

Job Type: Permanent, Full-Time

Working Pattern: On-site, London

Sector: Robotics / Artificial Intelligence / Machine Learning

Robotics Learning Engineer – Cutting-Edge Humanoid Robotics

Are you a Robotics Engineer with experience in deep learning, reinforcement learning or robot manipulation? Would you like to work on some of the most advanced humanoid robotics technology being developed today?

We're working with an ambitious humanoid robotics company in London that is developing next-generation robots for real-world industrial applications.

This is an opportunity to join a highly skilled engineering team working at the intersection of robotics, artificial intelligence, machine learning and physical hardware.

You’ll be responsible for developing and improving the learning systems that enable humanoid robots to interact with and manipulate objects in real-world environments.

This isn't purely a simulation or research position. You’ll be working directly with physical robots, taking machine learning models from training through to deployment and improving their performance on real hardware.

The Role

As a Robotics Learning Engineer, you’ll join a specialist Manipulation team developing intelligent robotic capabilities.

Your responsibilities will include:

• Developing and fine-tuning robot manipulation policies using behaviour cloning and reinforcement learning.
• Training, evaluating and deploying machine learning models onto real robotic hardware.
• Improving training datasets through preprocessing, curation and quality optimisation.
• Working with simulation engineers to develop reinforcement learning environments and improve sim-to-real transfer.
• Collaborating with data collection teams to improve training data quality, diversity and coverage.
• Developing new observation and action spaces to support advanced robotic manipulation.
• Working closely with teleoperation and controls engineers to improve robot movement, responsiveness and smoothness.
• Investigating hardware and control limitations and contributing to future robot design improvements.
• Running experiments, analysing results and identifying ways to improve robotic performance.

What We’re Looking For

We’re particularly interested in engineers who combine practical robotics experience with applied machine learning expertise.

Essential experience:

• At least 3 years' experience working with robotic systems in industry or research, with demonstrable project outcomes.
• Hands-on experience developing, testing or deploying software on real robots.
• Experience with deep learning and neural network training or fine-tuning.
• Strong Python programming skills and familiarity with PyTorch, JAX or similar frameworks.
• Understanding of robotic control systems, teleoperation or motion planning.
• Experience working with machine learning datasets, model training and evaluation.
• Strong software engineering skills, including debugging, testing and maintainable code development.
• Ability to analyse experimental results and solve complex technical problems.

Desirable experience:

• Reinforcement learning applied to robotic systems.
• Robot manipulation, grasping or dexterous control.
• Vision-Language-Action (VLA) models.
• Behaviour cloning and imitation learning.
• Sim-to-real reinforcement learning.
• Experience with OpenVLA, diffusion policies or similar robotic learning frameworks.
• Experience with distributed training, checkpointing and machine learning infrastructure.
• Research publications or open-source contributions in robotics or machine learning.

What’s On Offer?

• Competitive salary and meaningful equity participation.
• 30+ days of paid leave, including annual leave, bank holidays and company closure days.
• Private healthcare.
• Pension contributions.
• Free daily breakfast, catered lunch and office snacks.
• Opportunity to work directly with advanced humanoid robotics hardware.
• Collaboration with leading robotics engineers, AI researchers and technical specialists.
• Significant technical ownership and opportunities to influence future product development.
• A fast-moving engineering environment where your work directly impacts real-world robotic capabilities.

Why Apply?

This is an opportunity to work on technology that is moving beyond research laboratories and into real industrial environments.

You’ll be joining a team solving some of the hardest challenges in robotics, combining machine learning, control systems, simulation and real-world hardware to develop increasingly capable humanoid robots.

If you’ve worked on robotic learning, manipulation, reinforcement learning or the deployment of machine learning models onto physical robots, we’d be keen to hear from you.