Researcher - AI Computing System

About the role

Huawei Canada’s Advanced Computing and Storage Lab is seeking a researcher on a 12‑month contract to advance AI computing systems on the Ascend platform. The role focuses on large‑model training and inference, targeting performance, efficiency and usability improvements through low‑precision and multimodal techniques.

Key responsibilities

• Design and develop optimization solutions for AI training and inference, including FP8 optimization and reinforcement‑learning‑driven training agents.
• Build and maintain stable, efficient AI training clusters using dynamic configuration and precision control.
• Develop software frameworks, operator and acceleration libraries for NPU platforms to accelerate large‑model workloads.
• Drive innovation in low‑precision training, parallel strategy tuning, and reinforcement‑learning‑based resource optimization.

Required profile

• Ph.D. or Master’s degree in Computer Science, Computer Engineering or related fields.
• Familiarity with large‑model architectures such as DeepSeek and LLaMA and experience optimizing LLM, MoE, or multimodal models.
• Understanding of AI accelerator hardware (GPU/NPU) and ability to co‑optimize software and hardware components.

Required skills

• Proficient in Python, C and C++ programming.
• Experience with GPU and NPU platforms.
• Knowledge of low‑precision training techniques (e.g., FP8).
• Background in reinforcement learning and multimodal optimization.