About
This is an early-stage robotics company building the intelligence layer for industrial machines – forklifts, cranes, excavators and other heavy equipment.
Pure controls is too narrow for this problem. Pure ML is too. The challenge is getting large, safety-critical machines to move well in the real world, using the right mix of classical control, learned behaviour, and data from live operation.
This role sits in the locomotion and controls team, working on movement, navigation, and machine behaviour for autonomous industrial equipment. The company already has a remote-operation system in the field, which means the autonomy work is grounded in real operator data and real constraints.
The team wants engineers who can move across both worlds and know where each approach breaks.
What you'll do
* Improve machine control and behavioural performance across real operating conditions
* Build and tune control systems for locomotion and task execution
* Apply RL or other learning-based methods where they improve capability or robustness
* Design experiments that connect simulation, offline analysis, and on-machine validation
* Work closely with perception, navigation, and platform teams to improve full-system behaviour
* Analyse failure cases and turn them into better models, controllers, or data
* Help shape how learned and classical methods are combined in production
What you'll need
* Strong robotics autonomy experience in controls, planning, or machine behaviour
* Hands-on experience with reinforcement learning or other learning-based control methods
* Strong Python and/or C++ skills
* Ability to work across modelling, experimentation, and production deployment
* Comfort with real-world robotics constraints, not just simulation
* Strong ownership and the ability to operate with broad scope
Optional Bonus
- Experience with heavy equipment, off-road autonomy, or industrial robotics- Background in system identification, MPC, or sim-to-real methods
Shortlisted candidates will be contacted within 48 hours.