Machine Learning Engineer, Zoox

• 2+ years of industry experience, and/or PhD, developing neural rendering techniques like Gaussian Splatting, NeRFs, or 3D reconstruction
• 2+ years of experience developing software with Python and/or modern C++
• Expertise with machine learning frameworks PyTorch or Tensorflow
• Familiarity with 3D graphics algorithms, such as 3D geometry and camera models
• Strong mathematical skills and understanding of 3D linear algebra and probabilistic techniques

Desirable

• The 3D Simulation group at Zoox is looking for 3D Machine Learning engineers to simulate sensors (cameras, lidar, radar), combining GenAI/ML and modern 3D graphics techniques to close the gap between simulation and reality
• You will have access to the best sensor data in the world and an incredible infrastructure for testing and validating your algorithms
• Collaborate with Sensor Sim, Perception, and V&V teams to evaluate and iteratively improve the fidelity of sensor simulation
• The 3D Sensor Simulation team’s work is at the core of creating high fidelity synthetic data for training and testing AV AI, as well as real time sensor data for hardware-in-the-loop simulation
• Research, implement, and optimize state of the art 3D rendering of sensor data, leveraging GenAI/ML and 3D graphics
• Develop realism metrics with V&V to show measurable impact of your improved sensor fidelity
• Collaborate with Perception and Safety teams to improve realism of sensor simulation for high fidelity synthetic data
• Improve rendering and ML inference tooling for generating realistic data at scale