Description
Humanoid is hiring a Neural Network Performance Engineer for its VLA team in London. The role focuses on improving the performance of neural-network-based control policies for humanoid robots, including analyzing bottlenecks, adapting models to new hardware, implementing custom kernels, quantizing models, and modifying architectures. Candidates should have at least three years of deep-learning systems experience and one year of neural-network inference performance experience, along with strong Python, PyTorch/JAX, GPU, profiling, and debugging skills.
