ML Systems Software Development Engineer Intern, Annapurna Labs - 2027

The Toronto Neuron team at Amazon's Annapurna Labs is looking for driven engineering students ready to do serious work on custom silicon and machine learning systems . This isn't a fetch-coffee internship - you'll be embedded in a team that designs the chips and software powering AWS at global scale, working on ML Systems Software Development from day one in Toronto, Ontario .

Depending on your background and interests, you'll be placed on either the Frontier Model Performance team or the Developer Experience (DevEx) team . Both focus on AWS Trainium - Amazon's custom machine learning silicon - and both involve real engineering problems with real customer impact.

About the Role: ML Systems Software Development Engineer Intern On the Frontier Model Performance team , you'll take newly released machine learning models from first bring-up to peak performance on AWS Trainium silicon . That could mean writing and tuning kernels, optimizing sharding and model execution, building benchmark infrastructure, or tackling architectural bottlenecks that shape future chip designs. The best ideas get turned into reusable Neuron components and optimization techniques that flow into flagship open-source and customer workloads. On the DevEx team , you'll build the tools that make Trainium performance visible - owning Neuron Explorer and developing profiling, debugging, and analysis capabilities so engineers can identify bottlenecks and optimize faster. That could mean low-level performance data collection , analysis engines, interactive visualizations, or developer workflows.

In either team, you'll work alongside experienced machine learning and systems engineers who are invested in your growth. You'll collaborate across disciplines, take ownership of real deliverables, and contribute to software that runs at massive scale. Safety, code quality, and clear technical communication are valued throughout the team.

Benefits and Salary This internship position in Toronto, ON offers a starting salary of $100,810 CAD annually . Amazon also provides basic life and AD&D insurance , paid time off , and access to other resources aimed at improving health and well-being. Internship terms are flexible, with options for a 12-16 month placement starting May 2027 or a 3-4 month placement starting January 2027, May 2027, or September 2027.

Job Details Job Type: Internship

Company: Amazon (Annapurna Labs)

Location: Toronto, ON, Canada

Requisition ID: 10538066

Date Posted: September 11, 2026

Schedule: 3-4 month or 12-16 month internship terms available

Pay: $100,810 CAD Annually

Responsibilities Your day-to-day will vary depending on which team you join, but the common thread is solving hard engineering problems at the intersection of machine learning and custom hardware . You'll own meaningful workstreams, produce results that matter to real customers, and learn from engineers who've built some of the most sophisticated ML infrastructure on the planet.

Bring up and optimize state-of-the-art machine learning models for peak performance on AWS Trainium silicon

Write and tune kernels , optimize model sharding and execution, and tackle architectural bottlenecks that influence future hardware designs

Build benchmark and measurement infrastructure to evaluate model performance across hardware generations

Develop reusable Neuron components and optimization techniques applicable to flagship open-source and customer workloads

Build profiling, debugging, and analysis tools within the Neuron Explorer ecosystem to surface performance data for engineers

Design and implement low-level performance data collection pipelines, analysis engines, and interactive visualizations

Collaborate with ML and systems engineers to identify and resolve performance bottlenecks across model and kernel execution

Requirements / Skills Amazon is looking for engineering students who are genuinely curious about machine learning systems, compilers, or hardware-software co-design . You don't need to be an expert in everything - but you should have solid fundamentals, hands-on coding experience, and a demonstrated interest in at least two of the technical domains listed below.

Current enrolment in a Bachelor's degree or higher in Computer Science, Computer Engineering, Electrical Engineering, or a related field

Programming experience in Python, C, and/or C++ through coursework, research, or a previous internship

Academic, research, or project experience in at least two of: performance engineering or profiling; kernel or parallel programming; developer tooling; data structures and algorithms; ML frameworks such as PyTorch or JAX; compiler or ML systems technologies such as LLVM, MLIR, XLA, or TVM

Experience optimizing machine learning models or writing kernels for GPUs, ML accelerators, or FPGAs (preferred)

Ability to communicate technical challenges clearly and work independently through ambiguous or undefined problems

Full stack development experience including TypeScript, React, or Go is a plus for DevEx candidates

Currently enrolled in a Bachelor's degree program or higher in Computer Science, Computer Engineering, Electrical Engineering, or a related field; programming experience using Python, C, and/or C++; strong interest and academic, research, or project experience in at least two relevant technical areas.

#J-18808-Ljbffr