Senior Embedded AI Engineer - C/C++, Low-Power Inference

Qualcomm; Markham; On-site; CAD 98,640 - 164,400; Full time

Job summary

Qualcomm Canada ULC is seeking an engineer to deploy ML models on a low power AI accelerator. You will map high level operators to hardware instructions, optimize graphs, schedule tasks, plan memory and implement operators with quantization.

The role focuses on improving inference efficiency and accuracy on Qualcomm's hardware, requiring C/C++ proficiency, Linux/Android experience, and embedded hardware familiarity.

Qualifications

• Bachelor's degree with 2+ years of related work experience or higher; or Master's with 1+ year; or PhD.
• Master's degree preferred in a related field.
• Experience with ML frameworks (TensorFlow, PyTorch, etc.) is a plus.

Responsibilities

• Apply ML knowledge to extend training or runtime frameworks for efficiency.
• Architect and develop ML hardware co-designs for inference/training.
• Develop optimized software for ML models on Qualcomm hardware.
• Collaborate with team for joint design and development.
• Apply ML techniques to products/AI solutions for customers.
• Prototype complex ML algorithms with guidance from senior engineers.

Skills

C/C++

Performance optimization

Linux/Android development

Embedded hardware

ML knowledge

Education

Bachelor's degree in Computer Science/Engineering/IS

2+ years in Hardware/Software/Systems Eng

PhD in related field

Job description

Qualcomm Canada ULC is seeking an engineer to deploy ML models on a low power AI accelerator. You will map high level operators to hardware instructions, optimize graphs, schedule tasks, plan memory and implement operators with quantization.

The role focuses on improving inference efficiency and accuracy on Qualcomm's hardware, requiring C/C++ proficiency, Linux/Android experience, and embedded hardware familiarity.