Robotics Simulation Intern

What you will actually do

SimRoboX is the simulator our own research runs on, which means every rough edge you fix is one a researcher hits the next morning.

Depending on where you are strongest, you will work on:

• Environment authoring — making it straightforward to describe a robot and a task without fighting the framework
• Domain randomization — the parameter sweeps that decide whether a policy survives contact with the real world
• Evaluation tooling — the dashboards and metrics that tell you a policy is ready, and the honest ones that tell you it is not

What we are looking for

• Python, and enough familiarity with a physics engine to know why contacts are hard
• Some exposure to reinforcement learning — you should know what an episode and a reward function are
• Bonus: URDF, ROS, or any experience getting a learned policy onto real hardware

How we work

You will have a supervisor, a weekly one-to-one and a research log. Interns here have taken their work to a deployed hardware demo — twice.

Applying

Send a repo, a simulation you built, or a write-up of something that did not work and why. Curiosity reads better than credentials.