Senior ML Compiler Engineer - HW/SW Co-Design

Qualcomm; Markham; On-site; CAD 120,000 - 180,000; Full time

Job summary

Qualcomm Canada ULC in Markham is seeking a senior engineer to develop ML/AI compilers for efficient mappings of ML workloads on Qualcomm hardware. You will work on production and exploratory versions, collaborating with hardware and software teams.

The role emphasizes performance and power trade-offs, pre-silicon predictions, and close study of ML network trends, with hands-on work in C++, Python, and modern tooling.

Qualifications

• Bachelor's degree with 8+ years in hardware/software/systems.
• Proven ability to learn and adapt in fast-changing environments.
• Strong problem-solving and debugging skills.
• Experience with ML/AI compiler development preferred.

Responsibilities

• Develop and evolve ML/AI compilers for efficient mappings of ML workloads on HW.
• Analyze ML workloads to drive future ML HW/SW features.
• Define algorithms for compiling ML workloads to achieve high performance and low power.
• Collaborate across cross-site, cross-functional teams.
• Create performance-driven simulations (C++, Python) for HW/SW algorithms.
• Pre-silicon prediction of performance for ML algorithms.
• Investigate and propose improvements to tackle bottlenecks.

Skills

Problem solving

Analytical thinking

Strong communication

Detail oriented

Education

Bachelor's degree in Computer Science, Engineering, Information Systems, or related field

Master's degree in Computer Science, Engineering, Information Systems, or related field

PhD in Computer Science, Engineering, Information Systems, or related field

Tools

C++

Python

Docker

Git

Jenkins

clang/MSVC

PyTorch

ONNX

Job description

Qualcomm Canada ULC in Markham is seeking a senior engineer to develop ML/AI compilers for efficient mappings of ML workloads on Qualcomm hardware. You will work on production and exploratory versions, collaborating with hardware and software teams.

The role emphasizes performance and power trade-offs, pre-silicon predictions, and close study of ML network trends, with hands-on work in C++, Python, and modern tooling.