Intern Researcher - AI Agent Evaluation

Summary

Huawei Canada has an immediate 4-12 month internship opening for an Intern Researcher. The role focuses on researching and developing advanced AI technologies, specifically evaluating large language model-based agents to enhance their reliability and reasoning capabilities.

Responsibilities

• Contributing to cutting-edge research on the capabilities and evaluation of large language model-based agents to shape the next generation of reliable, reasoning-capable AI systems
• The role focuses on addressing fundamental questions that arise as AI agents are increasingly deployed in complex, long-horizon tasks, specifically regarding how they manage context, reason under uncertainty, and maintain reliable behavior
• Research strategies will involve drawing from formal reasoning, causal inference, and theoretical foundations to better characterize and understand agent behavior in various environments
• This is strictly a research-first position where you will be expected to formulate complex problems, run detailed experiments, and contribute to publishable work rather than building production systems
• The ultimate goal of the work is to develop principled methods and theoretical foundations that ensure meaningful progress in the field of agent evaluation and autonomous reasoning

Skills

• Currently enrolled in a Master or PhD program in Computer Science, Electrical Engineering, Mathematics, or a closely related field
• Demonstrated research output such as publications, preprints, or workshop papers at top-tier venues
• Solid foundation in machine learning and NLP fundamentals
• Demonstrated ability to work on a research project from hypothesis to result
• Clear and effective written and verbal communication skills
• Research experience in one or more of the following areas is an asset: LLM Architectures, causal inference and counterfactual reasoning, Formal Methods, or Neurosymbolic AI
• Experience designing, critiquing, or extending LLM reasoning or agent evaluation benchmarks

Qualifications

Must Haves

• Currently enrolled in a Master or PhD program in Computer Science, Electrical Engineering, Mathematics, or a closely related field
• Demonstrated research output such as publications, preprints, or workshop papers at top-tier venues
• Solid foundation in machine learning and NLP fundamentals
• Demonstrated ability to work on a research project from hypothesis to result
• Clear and effective written and verbal communication skills

Nice to Haves

• Research experience in one or more of the following areas is an asset: LLM Architectures, causal inference and counterfactual reasoning, Formal Methods, or Neurosymbolic AI
• Experience designing, critiquing, or extending LLM reasoning or agent evaluation benchmarks