Research Engineer - AI Workload & Systems

Huawei Canada has an immediate 12-month contract opening for a Researcher.

About the team:

The Computing Data Application Acceleration Lab aims to create a leading global data analytics platform organized into three specialized teams using innovative programming technologies. This team focuses on full-stack innovations, including software-hardware co-design and optimizing data efficiency at both the storage and runtime layers. This team also develops next-generation GPU architecture for gaming, cloud rendering, VR/AR, and Metaverse applications.

One of the goals of this lab are to enhance algorithm performance and training efficiency across industries, fostering long-term competitiveness.

About the job:

Frontier AI Technology Research

• Track the evolution of state-of-the-art AI model architectures, including Large Language Models (LLMs), Vision Language Models (VLMs), advanced attention mechanisms, and Mixture-of-Experts (MoE) architectures.
• Analyze the computational characteristics of emerging model architectures and Agentic AI training and inference workloads.

Hardware-Oriented Workload Analysis

• Develop a systematic framework to map AI applications and workloads to hardware requirements.
• Identify key computational patterns, design efficient inference deployment strategies, and build analytical performance models.
• Evaluate the impact of algorithmic and hardware innovations on system performance, and provide quantitative, explainable architectural recommendations for next-generation AI accelerators.

The total target annual compensation (based on 2,080 hours per year) for this position ranges from $127,000 to $225,000 depending on education, experience, and demonstrated expertise.

About the ideal candidate：

• Strong understanding of modern AI model architectures and emerging trends, including sparse attention, linear attention, Mixture-of-Experts (MoE), and related techniques.
• Solid knowledge of AI accelerator architectures (e.g., GPUs, TPUs), with deep understanding of memory hierarchy, interconnect technologies, and hardware performance bottlenecks.
• Hands-on experience with AI kernel development using technologies such as Triton, TileLang, CUDA, or equivalent. Strong understanding of FlashAttention and other state-of-the-art kernel optimization techniques.
• Experience with modern LLM inference systems and optimizations, including vLLM, SGLang, or similar serving frameworks. Familiarity with the internals of deep learning frameworks such as PyTorch and JAX.
• Experience using hardware performance analysis and profiling tools, with the ability to develop Roofline-based performance analysis methodologies and tools.
• Ph.D. in Artificial Intelligence, Computer Architecture, Computer Systems, or a closely related field is an asset.
• Demonstrated research contributions through publications or influential open-source projects in AI infrastructure, systems, or computer architecture is an asset.
• Experience deploying and optimizing large-scale AI training or inference systems in production environments is an asset.