NVIDIA is looking to accelerate the AI revolution by bringing general intelligence to humanoid robots. This role focuses on building a Neural Reconstruction Engine for production usage within the NuRec Project, aiming to solve complex, previously unsolved real-world problems in robotics.
Requirements
- Expert in C++ and Python.
- Experience with Neural Reconstruction, e.g. 3DGS, NERF etc.
- Experience with Deep Learning Research and Development.
- Proficient with Linux environment.
- Architected and built software stack for Robotics.
- Built large scale training model with deep learning framework e.g. tensorflow, PyTorch, tensorRT
- Experience accelerating algorithms with CUDA.
Responsibilities
- Work with multiple teams to bring prototypes up to the production system.
- Deploy algorithms, models and integrate them as a neural reconstruction pipeline.
- Research on SOTA methods for neural reconstruction.
- Build tools to evaluate performance of models and algorithms.
- Train models with Neural Reconstructed scenes and deploy them on robots.
Other
- MS or PHD in Computer Science or relevant field (or equivalent experience).
- 5+ years of experience working on systems and pipelines.
- Exceptional problem-solving, analytical and troubleshooting skills.
- Excellent written and verbal interpersonal skills.
- Experience in leading teams to build a complex software stack.
