ML Engineer - Simulation & Robotics (up to £100k)

Dex

Cambridge

On-site

GBP 70,000 - 100,000

Full time

20 hours ago
Be an early applicant

A complete application in a minute

Tailored resume and cover letter, ready to send.

Get past ATS filters

Job summary

Dex in Cambridge is building a team to develop GPU-accelerated physics simulation and training infrastructure for robot brains. You will join a veteran engineering group working on high-speed simulators and synthetic data pipelines to enable robust policy learning.

The role focuses on training policies with RL or imitation learning using PyTorch, shipping production-grade Python software, and solving challenging sim-to-real problems with hardware integration in Cambridge.

Qualifications

  • Experience training and evaluating policy models for robotic or embodied systems using PyTorch.
  • Production-grade Python engineering experience building, debugging, and extending ML/robotics software systems.
  • Deep practical exposure to at least one sim-to-real problem: physics simulation, domain randomisation, system identification, or robustness-driven training.
  • Proven ability to debug across the entire stack, isolating issues between simulator dynamics, ML pipelines, and real hardware behaviour.
  • Ability to work on-site with physical robot hardware in Cambridge, UK.

Responsibilities

  • Train robust robot policies using reinforcement learning, imitation learning, and related paradigms targeting high-reliability physical transfer.
  • Develop domain randomisation, system identification, and robustness-driven training pipelines to systematically shrink the sim-to-real gap.
  • Deploy learned policies directly to physical robot hardware and edge compute, diagnosing root causes where real-world dynamics diverge from simulation.
  • Build evaluation methods, automated benchmarking tooling, and synthetic data generation workflows to quantify policy performance at scale.
  • Partner closely with core simulation engineers to integrate trained models into the wider robotics infrastructure and drive hardware capabilities forward.

Skills

Reinforcement learning
Imitation learning
PyTorch
Python
Sim-to-real
Debugging across stack

Job description

This role is with one of Dex's trusted partner companies. We work closely with their teams to truly understand their culture, goals, and what they're looking for, so we can match you with the right opportunity and give you context about the role before you commit to a process.

The role

Founded by veteran engineers who spent over a decade building core physics engines at the world's leading GPU computing giant, this Cambridge-based team of 11 is eliminating the primary bottleneck in robotics: physical trial-and-error. Backed by top-tier VCs with a long runway, they are building ultra-fast GPU physics simulation and training infrastructure to generate synthetic data and train robot brains at unprecedented scale.

The work
  • Train robust robot policies using reinforcement learning, imitation learning, and related paradigms targeting high-reliability physical transfer.
  • Develop domain randomisation, system identification, and robustness-driven training pipelines to systematically shrink the sim-to-real gap.
  • Deploy learned policies directly to physical robot hardware and edge compute, diagnosing root causes where real-world dynamics diverge from simulation.
  • Build evaluation methods, automated benchmarking tooling, and synthetic data generation workflows to quantify policy performance at scale.
  • Partner closely with core simulation engineers to integrate trained models into the wider robotics infrastructure and drive hardware capabilities forward.
What You Bring
  • Demonstrated track record training and evaluating policy models (RL or imitation learning) for robotic or embodied systems using PyTorch.
  • Production-grade Python engineering experience building, debugging, and extending ML or robotics software systems, not just research notebooks.
  • Deep practical exposure to at least one sim-to-real problem: physics simulation, domain randomisation, system identification, or robustness-driven training.
  • Proven ability to debug across the entire stack, isolating issues between simulator dynamics, ML pipelines, and real hardware behaviour.
  • Ability to work on-site with physical robot hardware in Cambridge, UK.

As part of the recruitment process at Dex, we process your personal data in accordance with our Privacy Notice for Job Applicants. This notice explains how and why your data is collected and used, and how you can contact us if you have any concerns.

Get your free, confidential resume review.

or drag and drop your file here.

Similar jobs

Similar jobs worth comparing

Robotics ML Engineer: Real-World Simulation & RL (Cambridge)
Robotics ML Engineer: Real-World Simulation & RL (Cambridge)

Dex • Cambridge

On-site
GBP 70,000 - 100,000
Simulation Engine Engineer (up to £110k)
Simulation Engine Engineer (up to £110k)

Dex • Cambridge

On-site
GBP 90,000 - 150,000
High Performance Engineer (up to £140k)
High Performance Engineer (up to £140k)

Dex • Cambridge

On-site
GBP 90,000 - 120,000
Senior Machine Learning Engineer
Senior Machine Learning Engineer

Dex • Greater London

On-site
GBP 120,000 - 180,000
Senior Simulation Engineer - Robotics Manipulation Simulation
Senior Simulation Engineer - Robotics Manipulation Simulation

Secure Consult • Greater London

Hybrid
GBP 65,000 - 90,000
Research Engineer
Research Engineer

Dex • Greater London

On-site
GBP 90,000 - 150,000
Simulation Engineer - Manipulation
Simulation Engineer - Manipulation

Humanoid • Greater London

On-site
GBP 85,000 - 110,000
Stock options equity
30+ paid days off (UK)
Private healthcare
+4
Senior GPU Performance Engineer - Robotics Simulation (Onsite UK)
Senior GPU Performance Engineer - Robotics Simulation (Onsite UK)

Dex • Cambridge

On-site
GBP 90,000 - 120,000
Staff Machine Learning Engineer (£220,000 | RecSys | Python)
Staff Machine Learning Engineer (£220,000 | RecSys | Python)

Dex • Greater London

On-site
GBP 198,000 - 242,000
Competitive compensation
Performance bonus
Strong influence over technical direction
Product Engineer (Backend) - Senior/Staff (up to £180k)
Product Engineer (Backend) - Senior/Staff (up to £180k)

Dex • Greater London

Hybrid
GBP 110,000 - 170,000