Senior Autonomy Engineer · London

I build robots that see, learn and work in the real world.

Robotics and computer vision engineer. I take perception and robot learning from research to machines that run in the field, from heavy equipment to dexterous hands.

See the work →Work with me

Extend RoboticsHaptic-ACT policy plugging a connector

50% → 77%success plugging a connector with under 1 mm clearance, once operators could feel contactHaptic feedback · Extend Robotics
3% → 0.4%genuine users wrongly rejected by on-device face login, at one-in-a-million false acceptsVFace · Veridium
2.1×more correct matches in the auto-approved bucket, so human QC goes where it mattersUncertainty QC · Augmodo
100Krobot camera images turned into training data with zero hand labelsSelf-supervised spray targeting · Q-Bot

selected projects · robotics × vision × learning

Where perception and learning meet real machines.

All projects →

01Hive Autonomy·2026

RoboticsComputer vision

Simulation & perception for heavy machinery

Isaac Sim and MuJoCo twins of forklifts, wheel loaders and dozers, a fused camera-depth perception pipeline, and a see-through bucket view for operators.

target cut in truck-loading cycle: 10–15%
machine types simulated: 3

read case study →

02Extend Robotics·2024–25

RoboticsComputer visionMachine learning

Full-stack robot learning & Haptic-ACT

The full robot learning backend, from VR demonstration through AWS training and serving to real-time policies on Jetson Orin, for ACT, GR00T N1.6 and π0. Plus Haptic-ACT, force feedback for gentler manipulation (IEEE IROS 2025).

MSD plugging success with haptic feedback (teleop): 50% → 77%
lower avg fingertip force vs ACT, real robot: ~15%

read case study →

03Q-Bot·2021–22

RoboticsComputer visionMachine learning

Self-supervised spray targeting & auto-drive

A U-Net trained on years of operator behaviour, with no hand labels, that tells Q-Bot's Mk7 spray robot where to insulate next and lets it drive there on its own. First-author paper at IEEE MMAR 2022.