Senior Gpu And Ml Kernel Optimization Engineer

Advance machine learning acceleration through kernel optimization, GPU expertise, and parallel programming for Amazon's Neuron platform in Toronto.As a Senior GPU and ML Kernel Optimization Engineer, you will work at the hardware-software boundary within Amazon's Acceleration Kernel Library team. You will apply low-level optimization and machine learning knowledge to Inferentia and Trainium workloads. The position includes research, architecture, implementation, mentoring, and direct customer model support.Key Responsibilities:
Optimize kernels across generations of Neuron hardwareImplement fusion, sharding, tiling, and scheduling techniquesCreate metrics and automation for software efficienciesResolve root causes of software defects and bottlenecksParticipate in designs, code reviews, and stakeholder discussionsRequirements:
Five-plus years of software development experienceFive-plus years programming in at least one languageFive-plus years leading new or existing system architectureExperience with GPU kernel optimization or GPGPU computingKnowledge of parallel programming and optimization techniquesUse GPU, parallel programming, and kernel performance skills to improve machine learning workloads on Amazon accelerators.
#J-18808-Ljbffr