Train generative models and multi-task networks at scale, with millions of video clips and thousands of GPUs, with the objective to significantly enhance the capabilities of autonomous perception and planning
Take a pioneering role in the field of machine learning for vision by pushing the state of the art in large model scaling for vision and multi-modal applications
Explore the frontier of large-scale distributed training and optimization, deploying models to real customers all over the world
What You’ll Bring
A proven track record of successfully scaling and training large, complex models
Experience dealing with infrastructure challenges, possessing techniques for optimizing the training of high-capacity models efficiently
Proficiency in Python and a strong foundation in software engineering best practices
An in-depth understanding of the fundamentals of deep learning, including but not limited to knowledge of layer details, loss functions, and optimization techniques
Expertise in various neural network architectures for computer vision, speech, NLP, and related domains
Prior experience with PyTorch, TensorFlow, JAX, or similar deep learning frameworks
Familiarity with standard scientific computing primitives, including but not limited to batched operations, grid sampling, scatter/gathering, and einsum
Strong expertise in system-level knowledge, enabling you to profile and optimize training and evaluation processes efficiently, is preferred