AI Engineer, Model Pre-Training, Tesla AI

At Tesla AI, we are solving real-world autonomy at a scale that does not exist anywhere else. We are building the brains for millions of robots and to achieve this, we train frontier-scale foundation models on the world's richest, most diverse dataset of multimodal telemetry, vision, language, and robotic action. You will have access to unparalleled resources that set us apart from other companies in the AI industry. You will work with one of the richest real-world driving, robotics and human-interaction datasets available providing a unique, and perhaps the only, environment to investigate scaling laws for language, vision, multimodal, and sequential decision-making models under real product constraints for FSD, Optimus, and Digital Optimus. Tesla also offers one of the highest GPU resources per engineer in the industry, giving you the computational budget to run controlled scaling experiments, fit predictive laws, and train frontier-scale models far beyond typical research environments.

These resources enable you to pre-train foundation models and characterize their scaling behavior at a fidelity unmatched elsewhere, then close the loop by turning those insights into stronger base models for post-training and deployment on FSD, Optimus, and Digital Optimus in the real world.

• Design and iterate on pretraining recipes powering FSD (Full Self-Driving), Optimus, and Digital Optimus policy foundation models
• Perform scaling law analyses on model size, data size, data mixture, architecture, training compute, and other critical parameters to predict and optimize pre-training outcomes
• Develop and iterate on pre-training methodologies including data curation and mixture design, tokenization, objective design, and training recipes, informed by measured scaling behavior
• Run controlled experiments that isolate scaling variables, fit predictive laws, and translate findings into compute-optimal training plans and architecture choices (e.g., mixture of experts, hybrid attention)
• Drive data strategy for pretraining: curation, filtering, deduplication, domain mix, synthetic data, and quality signals that move evals and capabilities.
• Build and maintain infrastructure for efficient large-scale distributed pre-training and high-throughput experimental sweeps, resolving compute and memory bottlenecks end-to-end
• Evaluate pre-trained models with rigorous intermediate and downstream metrics; identify where scaling breaks (data quality, stability, saturation regimes) and close those gaps before models move into post-training
• Work closely with cross-functional teams across data, distillation, and deployment to ship stronger, compute-efficient base models into production, meeting stringent performance, safety, and reliability standards
• Contribute tools and frameworks that make pre-training and scaling analyses reproducible, measurable, and reusable across Tesla AI model families

• Deep, proven expertise in deep learning fundamentals, particularly in pre-training large-scale language, vision, or multimodal models
• Proven experience with scaling law analyses or compute-optimal training studies for large AI models, with a strong understanding of how model, data, and compute trade off
• Strong grasp of large-scale data pipelines, data mixture design, and the tradeoffs between model capacity, data quality/quantity, and compute
• In-depth knowledge of optimization at scale, training stability, loss/objective design, and modern neural architectures (mixture of experts, hybrid attention, and beyond)
• Hands-on familiarity with distributed training techniques for frontier-scale models (e.g., data/tensor/pipeline parallelism, sharded optimizers, mixed precision)
• Strong expertise in distributed computing and large-scale training infrastructure
• Proficiency in Python and a deep understanding of software engineering best practices
• Experience with deep learning frameworks such as PyTorch, TensorFlow, or JAX
• Demonstrated ability to work collaboratively in a cross-functional team environment
• Strong problem-solving skills and the ability to troubleshoot complex system-level issues across data, training, and evaluation

Along with competitive pay, as a full-time Tesla employee, you are eligible for the following benefits at day 1 of hire:

• Medical plans > plan options with $0 payroll deduction
• Family-building, fertility, adoption and surrogacy benefits
• Dental (including orthodontic coverage) and vision plans, both have options with a $0 paycheck contribution
• Company Paid (Health Savings Accounts) HSA Contribution when enrolled in the High-Deductible medical plan with HSA
• Healthcare and Dependent Care Flexible Spending Accounts (FSA)
• 401(k) with employer match, Employee Stock Purchase Plans, and other financial benefits
• Company paid Basic Life, AD&D
• Short-term and long-term disability insurance (90 day waiting period)
• Employee Assistance Program
• Sick and Vacation time (Flex time for salary positions, Accrued hours for Hourly positions), and Paid Holidays
• Back-up childcare and parenting support resources
• Voluntary benefits to include: critical illness, hospital indemnity, accident insurance, theft & legal services, and pet insurance
• Weight Loss and Tobacco Cessation Programs
• Tesla Babies program
• Commuter benefits
• Employee discounts and perks program

Pay offered may vary depending on multiple individualized factors, including market location, job-related knowledge, skills, and experience. The total compensation package for this position may also include other elements dependent on the position offered. Details of participation in these benefit plans will be provided if an employee receives an offer of employment.