Skip to main content

Machine Learning Performance Engineer, Annapurna Labs

amazonGranite City, ILLead L6
Overview: In this role you drive ML performance on AWS Neuron (Inferentia/Trainium) by developing high-performance kernels and profiling workloads across the software stack. You work closely with compiler, runtime, and hardware teams to close bottlenecks and improve the Neuron SDK for external developers. You’ll shape a new Tel Aviv performance engineering team, delivering tangible speedups and cost efficiency for customers at scale. This role offers broad ownership and meaningful impact on AI infrastructure.
Responsibilities: Design and implement high-performance compute kernels for ML operations leveraging Neuron architecture Profile end-to-end ML workloads to identify bottlenecks and drive measurable optimizations Enhance tooling and programming model for kernel/model developers, improving usability and debugging workflows Identify and drive optimization opportunities across the Neuron software stack (compiler, runtime, frameworks) Document software designs, runbooks, and performance findings for the broader team
Key requirements: 3+ years of non-internship professional software development experience Knowledge of Python and/or C++ programming Knowledge of computer architecture, operating systems, and parallel computing Experience with PyTorch, TensorFlow, and/or JAX ownership and accountability cross-functional collaboration problem solving and analytical thinking kernel writing or parallel programming (CUDA, Triton, CUTLASS, Pallas, Mojo, SIMD, MPI) compiler optimization or hardware-software co-design experience with ML frameworks (PyTorch, TensorFlow, JAX)
APPLY NOW (opens in a new tab)

Matching similar jobs

Level

Lead L6

Location

Granite City, IL

Occupation

Software Developers

Industry

Custom Computer Programming Services

Posted

today

APPLY NOW (opens in a new tab)