Sr Principal ML Optimization/GPU Architect

At Adobe, we’re driving our reinvention as an AI company and betting on Generative AI! Last year, we released many Generative AI capabilities under the Firefly umbrella – features used by more than 50% Photoshop users and generating 5B+ images, while doing AI responsibly and transparently! We are looking to bring on a Senior Principal ML GPU Architect to lead the ML GPU optimization team in Adobe Firefly, reporting to the Head of AI/ML and Data Platforms as a member of staff. You will partner with the Director of ML Engineering who is responsible for our platform engineering resources to unlock step function changes in training and inference speed/scale for all our ML workloads. This opportunity will not only enable you to make real world impact by optimizing ML workloads running on tens of thousands of GPUs, but also will enable you to have the opportunity to publish relevant work as either open-source or as technical publications in major conferences. The role involves hands on impact on all ML platforms powering inference, training, and data, as well as guiding the platform strategy towards higher scale and faster execution areas. What you’ll do Help drive ML Platform technical roadmap and Strategy. Lead and mentor highly motivated ML GPU optimization engineers/scientists. Write efficient forward and backward passes in CUDA/CuTe. Write optimized custom layers in Pytorch. Optimize ML training and inference code for large, distributed training/inference with FP8. Quality and performance analysis between data types such as BF16 and FP8 for large deep learning models. Understand and optimize H100 GPUs. Architect broader, end to end optimized training and inference code and schemes with Pytorch for large, distributed models. Write high quality, product level code that is easy to maintain and test following standard methodologies. What you'll need to succeed Proficiency in at least two of: Linux, Ansible, Docker, Kubernetes (7+yrs) Expert in Python and C++ Expert in CUDA/CuTe, NCCL, OpenCL, Triton Expert in Pytorch Experience with DDP, FSDP A minimum of seven years of experience in distributed computing A minimum of five years of experience working with AWS or similar cloud infrastructure Experience with HW resource management for ML training and/or deployment S., M.S., or Ph.D. in Computer Science, Computer Engineering or a related area Our compensation reflects the cost of labor across several U.S. geographic markets, and we pay differently based on those defined markets. The U.S. pay range for this position is $205,900 – $407,500 annually. Pay within this range varies by work location and may also depend on job‑related knowledge, skills, and experience. Your recruiter can share more about the specific salary range for the job location during the hiring process. Adobe is an equal opportunity employer. We hire hard‑working individuals, regardless of gender, race or color, ethnicity or national origin, age, disability, religion, sexual orientation, gender identity or expression, or veteran status. We know that when our employees feel appreciated and included, they can be more creative, innovative and successful. This is what it means to be Adobe For All. Learn more about our vision here. We will ensure that individuals with disabilities are provided reasonable accommodation to participate in the job application or interview process, to perform essential job functions, and to receive other benefits and privileges of employment. Please contact us to request accommodation. Adobe is proud to be an Equal Employment Opportunity and affirmative action employer. We do not discriminate based on gender, race or color, ethnicity or national origin, age, disability, religion, sexual orientation, gender identity or expression, veteran status, or any other applicable characteristics protected by law. Learn more. #J-18808-Ljbffr