Researcher – Web Agent

Huawei Canada; Markham; On-site; CAD 106,000 - 156,000; Full time

Job summary

Huawei Canada is seeking a Researcher to advance Web Rendering and Web Agents in the Web Technology Lab. The role focuses on evaluating how AI agents interact with real websites and how different web technologies affect performance, with a strong emphasis on hypothesis testing and rigorous documentation.

The ideal candidate will leverage AI-assisted coding to build prototypes, analyze LLM reasoning about web content, and contribute to studies that bridge AI agents with real-world web

Qualifications

• Solid understanding of how LLMs work, prompting, context management, tool use, reasoning patterns.
• Hands-on experience building or evaluating AI agents that interact with web, APIs, tools.
• Comfortable using AI-assisted coding tools to build prototypes independently.
• Strong research and writing skills with clear hypotheses and experimental design.

Responsibilities

• Research how state-of-the-art web agents interact with real-world websites, identifying failure modes and capability limits.
• Study how web tech stacks affect agent performance and define patterns to be more agent-friendly.
• Investigate how LLMs perceive and reason about HTML, rendered DOM, screenshots, and accessibility trees for different tasks.
• Prototype new architectures, interaction models, and agent frameworks to test hypotheses.

Skills

LLMs understanding

AI agents

Prototyping

Research writing

Job description

Huawei Canada has an immediate permanent opening for a Researcher.

About the team:

The Web Technology Lab focuses on two areas of next-generation web technology: Web Rendering and Web Agents. Web Rendering explores the Chromium rendering architecture, including the development of concurrent web layout systems and the use of vertical integration opportunities to optimize performance and user experience. Web Agents explores how AI agents understand websites, leverage web capabilities, and complete tasks for users. This includes browser-native agents and emerging technologies such as WebMCP, helping evolve the web from a space primarily designed for people to browse and interact with into one where people and AI agents can work together. Together, these efforts aim to make the web more efficient to render and easier for both people and AI agents to use.

About the job:

• Research how state-of-the-art web agents (browser-use, computer-use, MCP-based agents) interact with real-world websites — identifying failure modes, reliability gaps, and capability ceilings.
• Study how different web tech stacks affect agent performance, and define design patterns that make sites more agent-friendly.
• Investigate how LLMs perceive and reason about web content (HTML, rendered DOM, screenshots, accessibility trees) and which representations are most effective for different tasks.
• Prototype new architectures, interaction models, and agent frameworks to empirically test hypotheses and bridge Web capabilities with AI-driven Agents.

The total target annual compensation (based on 2,080 hours per year) ranges from $106,000 to $156,000 depending on education, experience, and demonstrated expertise.

Job requirements

About the ideal candidate:

• Solid understanding of how LLMs work — including prompting, context management, tool use, and reasoning patterns.
• Hands-on experience building or evaluating AI agents — particularly those that interact with external environments (web, APIs, tools).
• Comfortable using AI-assisted coding tools to build and iterate on prototypes independently; able to build prototypes independently.
• Strong research and writing skills, able to formulate clear hypotheses, design experiments, and communicate findings in structured written form.

Additional Information:

Huawei Canada is committed to a fair, inclusive, and accessible recruitment process. If you require accommodation during any stage of the hiring process, please let us know and we will work with you to meet your needs.

All applications for this position are reviewed directly by our hiring team, we do not use artificial intelligence tools to screen or select candidates.