Job Description

Salary: $151,300 - 261,500 per year Requirements:

• A minimum of 5 years of professional software development experience (non-internship)
• At least 5 years of programming experience in at least one software programming language
• 5 years of leadership in design or architecture of new and existing systems, including design patterns, reliability, and scaling
• 5 years of experience with the complete software development life cycle, covering coding standards, code reviews, source control management, build processes, testing, and operations
• Experience in mentoring, leading a technical team, or serving as a tech lead
• Preferred Qualifications:
• Bachelor's degree in computer science or a related field
• 6 years of total software development experience
• Expertise in accelerator architectures for machine learning (ML) or high-performance computing (HPC), such as GPUs, CPUs, FPGAs, or specialized architectures
• Experience in optimizing GPU kernels and GPGPU computing, including with CUDA, NKI, Triton, OpenCL, SYCL, or ROCm
• Proven experience with NVIDIA PTX and/or AMD GPU ISA
• Familiarity with developing high-performance libraries for HPC applications
• Proficient in low-level performance optimization for GPUs
• Experience in LLVM/MLIR backend development for GPUs
• Knowledge of ML frameworks (PyTorch, TensorFlow) and their GPU backends
• Experience with parallel programming and optimization techniques
• Understanding of GPU memory hierarchies and optimization strategies

• As a member of our Annapurna Labs team at Amazon Web Services (AWS), I will be involved in developing AWS Neuron, the software development kit designed to accelerate deep learning and GenAI workloads on our custom machine learning accelerators, Inferentia and Trainium. Working within the Acceleration Kernel Library team, I will focus on maximizing performance for AWS's ML accelerators. My duties will include designing and implementing high-performance compute kernels for ML operations by utilizing the Neuron architecture and programming models, analyzing and optimizing performance at the kernel level across various generations of Neuron hardware, and conducting detailed performance analysis using profiling tools to identify and resolve bottlenecks. I will collaborate closely with teams to develop innovative kernel optimization techniques and will work directly with customers to optimize their ML models on AWS accelerators.

• AI
• AWS
• Backend
• CUDA
• Hardware
• Support
• LLVM
• Machine Learning
• PyTorch
• TensorFlow
• Web
• Cloud
• Architect
• Backbone
• Flow
• GitHub

More:

I am proud to be part of a team that values diverse experiences and encourages individuals with varied backgrounds to apply, even if they do not meet all the qualifications outlined. At Amazon Web Services (AWS), we strive to maintain an inclusive team culture that celebrates our differences. We prioritize work-life balance and support mentorship and career growth for our team members. Our environment fosters knowledge sharing and gives everyone the opportunity to work on exciting and innovative projects at the forefront of machine learning and high-performance computing. We offer flexibility in working hours and a commitment to personal and professional well-being. If I need accommodations during the application and hiring process, there are resources available to ensure I have support. Overall, we are looking for passionate individuals who are ready to contribute to shaping the future of AI acceleration technology with us.

last updated 52 week of 2025

Job Tags

Similar Jobs