Research Engineer – Learning-Based Animation & Simulation

JR United Kingdom; Slough; Hybrid; GBP 70,000 - 100,000; Full time

Job summary

A leading tech company seeks ML Engineers specializing in reinforcement learning and simulation to drive innovation in intelligent animation. You'll develop control policies, implement new research, and work in a collaborative environment that values speed and autonomy. This hybrid position in London offers competitive compensation and equity, along with relocation support for exceptional candidates.

Benefits

Competitive salary

Deep equity with meaningful upside

Hybrid-first culture

Relocation support for exceptional candidates

Qualifications

• Proven experience in reinforcement learning for control, robotics, or animation.
• Strong coding ability in Python.
• Familiarity with simulation environments like Isaac Sim, MuJoCo, or Unity.

Responsibilities

• Develop and train RL-based control policies for robotic characters.
• Implement new research ideas and build robust systems on modern simulation engines.
• Collaborate closely with a team on iterative prototypes.

Skills

Reinforcement Learning

Python

Simulation Environments

Generative Methods

Collaboration

Tools

Isaac Sim

MuJoCo

Unity

PyBullet

Job description

Location: London (hybrid strongly preferred); remote in EU possible for exceptional candidates

Type: Full-time

Seniority: Open from postdoctoral level to senior/staff level

About the Position

Our client is reimagining how virtual agents move, behave, and interact. By combining real-time simulation, reinforcement learning, and generative models, they’re building a new foundation for intelligent animation in digital worlds; from characters to physically grounded agents.

This isn’t just an ML research role - you’ll be implementing novel algorithms, working directly in simulation, and tuning systems that will run in production pipelines. We’re looking for engineers who thrive in high-agency environments and want to push the edge of embodied intelligence.

What You'll Do:

• Develop and train RL-based control policies for humanoid or robotic characters in simulation
• Experiment with generative methods for motion synthesis, behaviour blending, and skill composition
• Build robust systems on top of modern simulation engines (Isaac Sim, MuJoCo, Unity, etc.)
• Implement new research ideas from papers, then go further and make them work
• Work closely with a small, tight-knit team of engineers and researchers shipping iterative prototypes

Experience Required:

• Proven experience in reinforcement learning for control, robotics, or animation
• Strong coding ability in Python and comfort with performance-aware development
• Familiarity with simulation environments (Isaac Sim/Lab, MuJoCo, Unity, PyBullet)
• Curiosity for generative methods (VAEs, diffusion, trajectory priors)
• Strong grasp of research from groups like Levine, Peng, Abbeel, Finn, or Mordatch
• Comfortable working across both high-level ML and low-level physics code
• Self-motivated, fast learner, collaborative mindset

Bonus Experience (Nice to Have)

• Experience with character animation pipelines
• Unity engine or game dev familiarity
• Experience deploying models in real-time or latency-sensitive environments

Why Join

• Join an early team building tech with massive creative and technical potential
• Tackle some of the hardest problems at the intersection of AI, physics, and animation
• Work in a culture that values speed, autonomy, and deep technical craft
• Competitive salary + deep equity with meaningful upside
• Hybrid-first culture in London, with remote flexibility for exceptional contributors

**Relocation support/sponsorship is viable for exceptional candidates who intend to relocate to London**