Software Engineer, AI Performance Modeling & Co Design

What to Expect

The AI co-design team is dedicated to developing and optimizing AI systems that can scale efficiently to thousands of compute nodes, enabling large-scale training, reinforcement learning at scale, and real-time inference for Autopilot and Optimus. Our goal is to push the boundaries of performance, power, and latency on our custom-designed AI-ASIC cluster, while delivering exceptional scalability and performance.

This unique position sits at the intersection of AI research, performance modeling, and system design. You will work closely with AI engineers to optimize model architectures, leveraging network topologies to efficiently parallelize and shard models on our cluster. Additionally, you will collaborate with our hardware co-design team to model and optimize future generation performance, ensuring seamless integration of AI models and algorithms with next-generation hardware.

What You'll Do

• Develop and apply performance modeling techniques to analyze and optimize AI system performance, including dataflow, mapping, and sharding of AI models

Design and implement collective and multi-device parallel computing algorithms and architectures to accelerate AI workloads

Develop and maintain network models and simulations to analyze and optimize AI system performance in distributed environments

Collaborate with cross-functional teams to integrate performance modeling and optimization techniques into AI system design and development workflows

Develop and maintain software tools and frameworks to support performance modeling, parallel computing, and networking development

Work closely with the hardware co-design team to model and optimize future generation AI-ASIC architectures, ensuring seamless integration of AI models and algorithms with next-generation hardware

Co-design and optimize AI systems for future generations, considering emerging trends and technologies, such as 3D stacked processors, photonic interconnects, and novel memory architectures

Develop and apply co-design methodologies to optimize AI system performance, power, and area, while ensuring scalability, reliability, and manufacturability

What You'll Bring

Bachelor's Degree in Computer Science, Computer Engineering, Electrical Engineering, proof of exceptional skills in related fields with practical software engineering experience, or equivalent experience

Experience in performance modeling, parallel computing, and networking, with a focus on AI systems

Proficiency in Python and C++

Experience with AI networks, such as CNNs, transformers, and diffusion model architectures, and their performance characteristics

Experience with parallel computing frameworks and libraries, such as MPI, NCCL, or Gloo

Understanding of GPU and TPU accelerators architecture, memory hierarchy, and ability to model performance and identify performance bottlenecks

Strong understanding of computer networks, including network protocols, topology, and performance modeling

Experience with SW / HW co-design and optimization

Strong analytical and problem-solving skills, with the ability to analyze complex systems and optimize their performance

Compensation and Benefits Benefits

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

2 medical plan options with $0 payroll deduction

Expected Compensation $132,000 - $390,000 / annual salary + cash and stock awards + benefits

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.