Machine Learning Performance Engineer (AWS Neuron)

About the role

Join the AWS Neuron team in Tel Aviv and help shape the performance of next‑generation ML workloads on Amazon’s custom accelerators, Inferentia and Trainium. You will work at the intersection of machine learning, systems, and compiler technology to deliver best‑in‑class inference and training performance.

Key responsibilities

• Analyze high‑performance ML workloads running on Annapurna hardware.
• Develop and optimize kernels for critical ML operations.
• Enhance the Neuron SDK to improve developer experience and system capabilities.
• Collaborate with compiler, framework, and hardware teams to maximize end‑to‑end performance.
• Contribute to instruction scheduling, memory management, parallelism, and compiler enhancements.

Required profile

• Strong background in performance engineering for ML systems.
• Experience working with low‑level software stacks and hardware accelerators.
• Ability to solve complex, novel technical problems across the full software stack.
• Excellent collaboration skills across diverse engineering teams.

Required skills

• AWS Neuron SDK
• Inferentia and Trainium accelerators
• Kernel optimization
• Compiler development
• Instruction scheduling
• Memory management
• Parallelism techniques

What we offer

• Opportunity to work on cutting‑edge AI infrastructure.
• Startup‑like environment within a global tech leader.
• Impactful projects that serve a large, worldwide customer base.