We are seeking Interns in the following AI disciplines:
Train large-scale foundation and generative models that are optimized for performance and latency
Improve data engine for large scale and high-quality dataset curation
Reinforcement Learning for instilling objectives and improving overall robustness
Design compound AI systems for better planning and reasoning
What You’ll Do
Applied research in the areas of Foundation Models, including but not limited to computer vision, large language models and generative modeling
Work on cutting-edge techniques in AI - multi-task learning, video networks, multi-modal generative models, imitation learning, reinforcement learning, semi-supervised learning, self-supervised learning
Explore and implement novel AI tooling and techniques for efficient training and fine-tuning of large-scale models
Leverage millions of miles of driving data and interventions to build a robust and scalable end-to-end learning based self-driving system
Collaborate with a team to apply research findings to real-world challenges, ensuring high-quality system integration within existing platforms
Experiment with data generation and network driven data collection approaches to enhance the diversity and quality of training data
Ship production quality, safety-critical software to the entirety of Tesla’s vehicle fleet
What You’ll Bring
Demonstrated experience in machine learning frameworks and models such as PyTorch, TensorFlow, GPT, CNNs, and generative models
Strong experience with Python and software engineering best practices
Experience with one or more of imitation Learning, reinforcement learning (offline/off-policy), modern neural network architectures (e.g., GPT, diffusion, generative models), or related techniques
An “under the hood” knowledge of deep learning: layer details, loss functions, optimization, etc
Prior experience with sparse training techniques, neural network pruning, and generative modeling
Experience with training large models on distributed computing
Ability to work on complex problems and produce significant research and/or experience deploying production ML models at scale
Proven track record of innovations and executions in deep learning, demonstrated with shipping products or first-author publications at leading AI conferences