Ai Research Scientist, Generative Modeling, Autopilot Ai

You will play a critical role in our small, specialized team of machine learning scientists, specializing in Generative Modeling. Your primary focus will be to develop, train, and optimize large generative models that comprise the core world-understanding and decision-making of Autopilot. You will have access to one of the world’s largest training clusters to develop and scale our leading-edge models.

• A demonstrable track record in developing and training large generative models. Your experience should cover various aspects, including infrastructure challenges, optimizing the efficiency of the training process, and tuning the model for best performance.
• Proficiency in Python and a strong foundation in software engineering best practices. Your code should be maintainable, efficient, and well-documented.
• A deep understanding of deep learning fundamentals, encompassing in-depth knowledge of layer details, loss functions, and optimization techniques.
• Proficiency in implementing neural network architectures tailored for generative modeling.
• Experience with deep learning frameworks such as PyTorch, TensorFlow, JAX, or similar platforms.
• Familiarity with standard scientific computing primitives, including batched operations, grid sampling, scatter/gathering, and einsum.
• (bonus) Strong expertise in system-level knowledge, enabling you to optimize the training and evaluation processes efficiently and effectively.