Responsibilities
- Develop and implement machine learning models to improve the realism of simulated sensor data, including for example approximation of light transport, material models, 3D representation and black box sensor characterization.
- Develop content generation systems for creating diverse and realistic simulation scenarios, including perception signals from geospatial data, measurements from novel hardware, and structured and unstructured text conditioning.
- Create behavior models for various actors in simulations such as pedestrians, vehicles, drones.
- Design and implement reinforcement learning algorithms for training robotic systems within the simulation environment.
- Optimize ML model performance for simulation applications, including low latency inference, architectural improvements and model compression.
Requirements
- Master's or PhD in a relevant field (e.g. computer science, mechanical engineering, physics, math, robotics, etc.) or equivalent working experience.
- Nice to have: C
- Nice to have: C++
- Nice to have: Rust
- Nice to have: CUDA