ML Systems Engineer Intern: Performance & Profiling

Socket.dev; Toronto; On-site; CAD 86,000 - 116,000; Full time

Job summary

Amazon in Toronto, Ontario is seeking a Software Development Engineer Intern to help bring up and optimize state-of-the-art machine learning models for peak performance on AWS Trainium. You will build profiling tools and contribute to industry-leading developer workflows.

You will join the Toronto Neuron team (Annapurna Labs) and work with mentors across ML and systems engineering to tackle real customer challenges, gaining hands-on experience with cutting-edge silicon, compilers, and runtimes.

Qualifications

• Currently enrolled in a Bachelor's degree program in Computer Science, Computer Engineering, Electrical Engineering, or related field
• Proficient in Python, C, and/or C++
• Strong interest in performance engineering, kernel development, or ML systems technologies

Responsibilities

• Join the Toronto Neuron team to bring up and optimize state-of-the-art machine learning models for peak performance on AWS Trainium
• Build and improve profiling tools to help engineers identify bottlenecks and fix issues
• Work on kernel/low-level optimizations and developer tooling as part of a cross-functional team
• Collaborate with mentors across machine learning and systems engineering to solve real customer problems

Skills

Python

C

C++

Performance engineering

Kernel development

Education

Bachelor's degree in Computer Science/Computer Engineering/Electrical Engineering or related field

Tools

LLVM

MLIR

XLA

TVM

Job description

Amazon in Toronto, Ontario is seeking a Software Development Engineer Intern to help bring up and optimize state-of-the-art machine learning models for peak performance on AWS Trainium. You will build profiling tools and contribute to industry-leading developer workflows.

You will join the Toronto Neuron team (Annapurna Labs) and work with mentors across ML and systems engineering to tackle real customer challenges, gaining hands-on experience with cutting-edge silicon, compilers, and runtimes.