AI GPU Performance Engineer

WHAT YOU DO AT AMD CHANGES EVERYTHING We care deeply about transforming lives with AMD technology to enrich our industry, our communities, and the world. Our mission is to build great products that accelerate next-generation computing experiences – the building blocks for the data center, artificial intelligence, PCs, gaming and embedded. Underpinning our mission is the AMD culture. We push the limits of innovation to solve the world’s most important challenges. We strive for execution excellence while being direct, humble, collaborative, and inclusive of diverse perspectives. AMD together we advance_ THE ROLE: We are building IREE as an open-source compiler and runtime solution to productionize ML on a variety of usage scenarios and hardware targets. Among them, having wide and performant GPU support is critical. We aim at a broad range of GPU coverage, from mobile to datacenter, via a unified software stack. And we want an end-to-end solution from ML framework integration down to generating efficient kernels. There will be no short of intriguing technical challenges to tackle, and there are abundant chances to collaborate with industry experts working at different layers of the stack. If this sounds interesting to you, please don’t hesitate to reach out to us! THE PERSON: An ideal candidate should be familiar with parallel programming, GPU architectures, compilers, and/or high-performance kernels. He/she should be comfortable at performing quantitative analysis of workload and drive improvements at suitable layers. Most importantly, the candidate is willing to learn and work across boundaries. KEY RESPONSIBILITIES: Develop and maintain GPU related compiler backends in IREE, e.g., LLVMGPU for ROCm/CUDA, or SPIR-V for Vulkan/Metal, etc. Quantitively analyze end-to-end model performance, identify bottlenecks, propose ideas to improve, prototype and productionize solutions. Profiling and improving kernel performance, e.g., matmul, matvec, reduction, etc. Follow industry directions and adopt emerging technologies in compilers, e.g., FlashAttention, etc. Design and implement Python interactions with compiler components, e.g., Triton Drive towards general solutions that benefit different GPU targets and the overall community Build principled and reusable compiler abstractions using MLIR at the proper level PREFERRED EXPERIENCE: Familiarity with MLIR, IREE, or LLVM Familiarity with ML compiler techniques like tiling, distribution, vectorization, etc. Deep understanding of GPU architectures and programming models Deep experience with writing high performance GPU kernels and GPU performance tuning Familiar with whole GPU software stack, from high level libraries down to driver details Experience debugging cross-stack issues and reducing user problems to actionable execution. Experience with GPU debugging and profiling tools Open-source development ethos PREFERRED ACADEMIC CREDENTIALS: Bachelor’s or Master's degree in Computer Science, Computer Engineering, Electrical Engineering, or equivalent #LI-G11 #LI-HYBRID AMD does not accept unsolicited resumes from headhunters, recruitment agencies or fee based recruitment services. AMD and its subsidiaries are equal opportunity employers and will consider all applicants without regard to race, marital status, sex, age, color, religion, national origin, veteran status, disability or any other characteristic protected by law. EOE/MFDV At AMD, your base pay is one part of your total rewards package. Your base pay will depend on where your skills, qualifications, experience, and location fit into the hiring range for the position. You may be eligible for incentives based upon your role such as either an annual bonus or sales incentive. Many AMD employees have the opportunity to own shares of AMD stock, as well as a discount when purchasing AMD stock if voluntarily participating in AMD’s Employee Stock Purchase Plan. You’ll also be eligible for competitive benefits described in more detail here. AMD does not accept unsolicited resumes from headhunters, recruitment agencies, or fee-based recruitment services. AMD and its subsidiaries are equal opportunity, inclusive employers and will consider all applicants without regard to age, ancestry, color, marital status, medical condition, mental or physical disability, national origin, race, religion, political and/or third-party affiliation, sex, pregnancy, sexual orientation, gender identity, military or veteran status, or any other characteristic protected by law. We encourage applications from all qualified candidates and will accommodate applicants’ needs under the respective laws throughout all stages of the recruitment and selection process.

THE ROLE: We are building IREE as an open-source compiler and runtime solution to productionize ML on a variety of usage scenarios and hardware targets. Among them, having wide and performant GPU support is critical. We aim at a broad range of GPU coverage, from mobile to datacenter, via a unified software stack. And we want an end-to-end solution from ML framework integration down to generating efficient kernels. There will be no short of intriguing technical challenges to tackle, and there are abundant chances to collaborate with industry experts working at different layers of the stack. If this sounds interesting to you, please don’t hesitate to reach out to us! THE PERSON: An ideal candidate should be familiar with parallel programming, GPU architectures, compilers, and/or high-performance kernels. He/she should be comfortable at performing quantitative analysis of workload and drive improvements at suitable layers. Most importantly, the candidate is willing to learn and work across boundaries. KEY RESPONSIBILITIES: Develop and maintain GPU related compiler backends in IREE, e.g., LLVMGPU for ROCm/CUDA, or SPIR-V for Vulkan/Metal, etc. Quantitively analyze end-to-end model performance, identify bottlenecks, propose ideas to improve, prototype and productionize solutions. Profiling and improving kernel performance, e.g., matmul, matvec, reduction, etc. Follow industry directions and adopt emerging technologies in compilers, e.g., FlashAttention, etc. Design and implement Python interactions with compiler components, e.g., Triton Drive towards general solutions that benefit different GPU targets and the overall community Build principled and reusable compiler abstractions using MLIR at the proper level PREFERRED EXPERIENCE: Familiarity with MLIR, IREE, or LLVM Familiarity with ML compiler techniques like tiling, distribution, vectorization, etc. Deep understanding of GPU architectures and programming models Deep experience with writing high performance GPU kernels and GPU performance tuning Familiar with whole GPU software stack, from high level libraries down to driver details Experience debugging cross-stack issues and reducing user problems to actionable execution. Experience with GPU debugging and profiling tools Open-source development ethos PREFERRED ACADEMIC CREDENTIALS: Bachelor’s or Master's degree in Computer Science, Computer Engineering, Electrical Engineering, or equivalent #LI-G11 #LI-HYBRID AMD does not accept unsolicited resumes from headhunters, recruitment agencies or fee based recruitment services. AMD and its subsidiaries are equal opportunity employers and will consider all applicants without regard to race, marital status, sex, age, color, religion, national origin, veteran status, disability or any other characteristic protected by law. EOE/MFDV

At AMD, your base pay is one part of your total rewards package. Your base pay will depend on where your skills, qualifications, experience, and location fit into the hiring range for the position. You may be eligible for incentives based upon your role such as either an annual bonus or sales incentive. Many AMD employees have the opportunity to own shares of AMD stock, as well as a discount when purchasing AMD stock if voluntarily participating in AMD’s Employee Stock Purchase Plan. You’ll also be eligible for competitive benefits described in more detail here. AMD does not accept unsolicited resumes from headhunters, recruitment agencies, or fee-based recruitment services. AMD and its subsidiaries are equal opportunity, inclusive employers and will consider all applicants without regard to age, ancestry, color, marital status, medical condition, mental or physical disability, national origin, race, religion, political and/or third-party affiliation, sex, pregnancy, sexual orientation, gender identity, military or veteran status, or any other characteristic protected by law. We encourage applications from all qualified candidates and will accommodate applicants’ needs under the respective laws throughout all stages of the recruitment and selection process.

Tags: No, USD $169,400.00/Yr., USD $242,000.00/Yr., US Careers (External)