Robotics Jobs in London

• London
• On-site
• Permanent

Builds action-conditioned generative world models that predict future video, proprioception, contacts and outcomes for humanoid robots, using diffusion and transformer architectures. The role spans large-scale distributed training, learned-simulator policy evaluation, synthetic rollout generation and sim-to-real transfer, working hands-on with real fleet data alongside pretraining, RL and manipulation teams.

PyTorch · JAX · Isaac Sim +1

• London
• On-site
• Mid-level
• Permanent

Train language-vision conditioned manipulation policies using reinforcement learning in both simulation and the real world, building task suites and real-world RL pipelines. Work closely with teleoperations and testing teams to collect trajectories and transfer policies from simulation to physical robots. Requires deep-learning experience with Python and PyTorch/JAX, plus hands-on work with LLMs, VLMs or generative models.

Isaac Sim · MuJoCo · Ray +2

• London
• Hybrid
• Lead
• Permanent

Lead the technical architecture integrating Wayve's autonomous driving software with automotive OEM and Tier 1 vehicle platforms, spanning sensors, middleware, ECU constraints, timing, diagnostics and safety. Define interface contracts, ownership boundaries and requirements processes, and represent Wayve in partner technical discussions while mentoring architects across teams.

CAN · FlexRay · Automotive Ethernet +2

• London
• Hybrid
• Senior
• Permanent

Owns the technical architecture and roadmap for LiDAR and cross-sensor 3D sensing, leading development of target-based and targetless calibration, LiDAR-to-camera registration, temporal alignment and motion compensation using LiDAR odometry, nonlinear least-squares optimisation and factor graphs. A senior individual-contributor role combining algorithm development, evaluation methodology and productionisation, with technical leadership and mentorship across the 3D sensing stack.

GTSAM · Ceres Solver · g2o +1

• London
• Hybrid
• Senior
• Permanent

Owns the end-to-end release cycle for AI driving models, coordinating ML engineering, evaluation, data science, platform, testing and product teams. Defines release gates and promotion criteria from training through validation to fleet deployment, drives regression investigations and rollback decisions, and builds tooling and KPIs to remove bottlenecks in a non-deterministic ML environment.

GitHub · Buildkite · CLI

• London
• Hybrid
• Mid-level · manages people
• Permanent

Lead a London-based team building the on-vehicle experiment platform that lets researchers deploy driving models to vehicles, run experiments and collect training data. Own people, roadmap and OEM partnerships while staying hands-on with embedded Linux, C++/Python and over-the-air deployment across vehicle fleets.

ROS 2 · DDS · QNX +4

• London
• Hybrid
• Lead
• Permanent

Lead the technical direction for an offline 3D reconstruction system that turns raw vehicle sensor data into quality-assessed 3D ground-truth geometry and ego motion. The role spans SLAM, lidar point-cloud registration, state estimation and sensor fusion, with responsibility for architecture, roadmaps, prototypes and cross-functional alignment across perception, calibration and validation teams.

ROS · PCL · OpenCV +3

• London
• Hybrid
• Lead · manages people
• Permanent

Lead a small team iterating on Wayve's end-to-end L2 driving model, balancing hands-on Python coding and experiments with directing priorities. You diagnose on-road issues, root-cause them, and drive fixes through data recipes, model architecture and evaluation, working across teams on model and vehicle interface problems.

Python

• London
• Hybrid
• Senior
• Permanent

Lead the technical architecture and roadmap for onboard localisation, designing and productionising state-estimation systems that fuse GNSS, IMU and wheel odometry for accurate vehicle pose estimates. The role spans algorithm design, prototyping, validation and deployment on real-time vehicle platforms, with hands-on technical leadership and mentorship across robotics, calibration, controls and sensor teams.

GTSAM · Ceres · g2o +7

• London
• Hybrid
• Senior
• Permanent

Hands-on role optimising inference for edge accelerators and GPUs so large transformer models run efficiently on low-cost, low-power in-vehicle compute. You'll profile model graphs, compilers, runtimes and kernels, implement optimisations like fusion, scheduling and custom kernels, and build benchmarking and regression tests across NVIDIA Orin/Thor and Qualcomm targets, working closely with model, platform and deployment teams.

TensorRT · CUDA · Qualcomm QNN +7

• London
• Hybrid
• Senior
• Permanent

Builds and evolves the data pipeline and orchestration platform behind Wayve's autonomous driving data, revamping the existing Harness system and creating a new DAG-based workflow platform. Works on Kubernetes-deployed, distributed infrastructure handling hundreds of parallel pipeline steps, batch inference and evaluation workflows, closely with partner teams.

Airflow · Dagster · Flyte +7

• London
• Hybrid
• Senior
• Permanent

Builds and ships computer vision and 3D perception models for driver assistance, working across the full ML lifecycle from data generation to training, evaluation and iteration. The role spans latency-constrained on-car perception and large-scale offline work using lidar, maps and extra compute to power an auto-labelling data flywheel.

LiDAR

• London
• Hybrid
• Lead
• Permanent

Owns the autonomy performance measurement framework for a public-scale robotaxi deployment, defining metrics, dashboards and release gates for an end-to-end learned driving system with no existing playbook. Runs the performance improvement loop across model, robot software, backend and evaluation tooling, leading cross-functional execution without direct reporting authority while staying hands-on in Python, C++ and SQL.

Databricks · Looker · Grafana +1

• London
• Hybrid
• Senior
• Permanent

Lead large-scale data enrichment and curation for the Core Model Safety team, mining hard and rare events and building workflows across pre/mid-training, reward modelling and self-supervised labelling. Design and train out-of-distribution detection mechanisms for robust driverless operation, working at the intersection of model engineering, infrastructure, compute and data.

PyTorch · CUDA

• London
• Hybrid
• Lead
• Permanent

A cross-functional individual contributor role shaping software architecture across onboard software, AI, simulation, training, inference and Robotaxi systems. The architect defines boundaries, interfaces and abstractions spanning AI, compute and vehicle platforms, leading large multi-team architectural changes and acting as a technical decision-maker on safety-critical systems.