Develop GPU compiler software for Intel AI platforms in Toronto, combining systems programming with modern compiler infrastructure under a hybrid work model. Contribute to technologies enabling productive machine learning framework execution.
As an Experienced Hire Compiler Engineer, you will build software for Triton, domain-specific languages, and emerging AI programming models. The work includes compiler front-end, middle-end, and code-generation contributions, alongside runtime integration and open-source participation. You will help address performance, portability, programmability, and scalability across heterogeneous hardware architectures.
Key Responsibilities: Design compiler features for AI programming models and DSLs
Test and debug software supporting high-performance GPU kernels
Create analyses and transformations that improve application performance
Work with architects, compiler engineers, and framework developers
Learn emerging AI software and heterogeneous computing technologies
Requirements: Relevant bachelor's degree plus one year of experience
Master's degree with zero years or PhD accepted
Strong C++ and/or Python programming background
Debugging, problem-solving, and systems concepts familiarity
Interest in performance optimization or programming languages
Bring systems programming and GPU performance interests to Intel's compiler engineering work supporting AI platforms in Toronto.