AI/ML Engineer

About the role

You will own how autonomous vehicles perceive and understand their surroundings, building the perception layer that turns raw sensor data into a reliable environmental model even in harsh outdoor conditions. This hands‑on position offers high ownership and expects you to get code running on real vehicles early.

Key responsibilities

• Design, build and deploy the perception stack across camera, LiDAR and radar inputs.
• Develop object detection, tracking and segmentation that operate reliably in dusty, glare‑filled, high‑heat environments.
• Build localisation and mapping capabilities for GPS‑denied and unmapped areas.
• Fuse multi‑sensor data into a unified, real‑time model for the planning stack.
• Optimise models for on‑vehicle inference on edge GPU hardware within strict latency budgets.
• Construct data pipelines for collection, labelling, validation and retraining from fleet data.

Required profile

• 3+ years building perception systems that have run on real robots or vehicles.
• Strong computer‑vision expertise in detection, tracking and segmentation on live sensor feeds.
• Hands‑on experience with SLAM, point‑cloud processing and multi‑sensor calibration.
• Experience with sensor fusion (Kalman filtering, factor graphs or learned fusion approaches).
• Proficiency in Python and C++ and deploying models to edge or embedded GPU hardware.
• Comfort with ROS/ROS 2 and field testing.
• Ownership mindset and ability to take features from test‑track to deployment.

Required skills

• Python
• C++
• ROS
• ROS 2
• SLAM
• Point cloud processing
• Multi‑sensor calibration
• Sensor fusion
• Kalman filtering
• Factor graphs
• Learned fusion approaches
• Object detection
• Tracking
• Segmentation
• Localisation
• Mapping
• Edge GPU hardware
• Model compression (quantisation, pruning, TensorRT)

What we offer

• Opportunity to build technology that runs on real vehicles in real missions.
• Early‑stage ownership in a fast‑growing AI and robotics startup.