“This role offers a unique opportunity to contribute to cutting-edge AI research in Trust and Safety, with a particular focus on understanding and evaluating the risks of autonomous AI systems. You’ll collaborate with researchers and practitioners to advance the science of AI safety and translate research into practical tools and approaches that support safer AI development and adoption. There is so much meaningful work ahead, and we’re eager to build it together.”
- Alyssa Lefaivre Škopac | Director, AI Trust and Safety About Amii One of Canada’s three main institutes for artificial intelligence (AI) and machine learning, our world-renowned researchers drive fundamental and applied research at the University of Alberta (and other academic institutions), training some of the world’s top scientific talent. Amii is a core partner in the Canadian AI Safety Institute (CAISI). Our cross-functional teams work collaboratively with Alberta-based businesses and organizations to build AI capacity and translate scientific advancement into industry adoption and economic impact.

About the Role
The Applied Research Scientist will advance the understanding and safety of increasingly autonomous AI systems through research on agentic AI. The role will investigate key safety challenges, develop rigorous evaluation methods, and explore approaches to improving system reliability, robustness, and oversight. Reporting to the Director of AI Trust and Safety, the Applied Research Scientist will lead research to understand, evaluate, and mitigate risks associated with agentic AI systems. As AI systems evolve from responding to individual prompts to independently planning, making decisions, using tools, and executing complex tasks, new safety challenges are emerging.



This role will advance scientific understanding of these challenges and develop practical approaches to addressing them. Research will build on Amii’s strengths, address critical gaps in the AI safety landscape, and be informed by national and international research priorities. Outputs will include academic papers, whitepapers, conference presentations, open-source tools, and evaluation methodologies. The position focuses on achieving excellence in three main accountabilities:
- Trust & Safety AI Research Projects
- Project Execution Potential research focus areas may include:
- Agentic AI safety evaluations: Developing rigorous methodologies, benchmarks, and testing environments to assess the capabilities, risks, robustness, and reliability of increasingly autonomous AI systems, including their behaviour across complex tasks and environments.
- Goal misgeneralization and unintended behaviours: Investigating how agentic systems interpret and pursue objectives, including failures where behaviours diverge from intended goals, instructions, or constraints.
- Oversight, monitoring, and control: Exploring approaches to evaluating and improving human oversight, monitoring, intervention, and control of autonomous AI systems.
- Emerging agentic safety risks: Investigating novel safety challenges associated with agentic AI, including failures arising from tool use, long-horizon tasks, interactions with external environments, and increasingly complex agent architectures.



These focus areas are intended to guide, rather than limit, the research agenda. We welcome candidates who bring new perspectives, methodologies, and research questions that could meaningfully advance the science of agentic AI safety. Required Skills / Expertise

Key Responsibilities
- Develops and communicates a transparent vision for how research projects contribute to the goals of the T&S; program at Amii, which may include applied industry projects
- Proposes, develops and executes Trust & Safety research proposals, defining objectives, scope, methodologies, evaluation approaches, and resource needs, with a particular focus on agentic AI safety.
- Develops and implements research experiments, evaluation pipelines, and environments, working with engineering colleagues where appropriate
- Identifies opportunities throughout the research process to translate findings into practical evaluation tools, methodologies, frameworks, and applications
- Authors and publishes peer-reviewed papers, whitepapers, and other scientific outputs
- Evaluates results and emerging insights to refine research plans and determine whether to continue, pivot, or discontinue projects based on research findings, new insights, technologies, or shifts in industry trends
- Represents the organization at conferences, workshops, and industry events, sharing knowledge and building a reputation as a thought leader in AI T&S;
- Mentors junior team members, providing technical guidance and strategic advice on T&S; projects
- Collaborates with cross-functional teams, coordinates project timelines, and resolves challenges to ensure project success
- Supports collaborations with Canadian and international research institutions to expand the s

📌 Applied Research Scientist (Edmonton)
🏢 Alberta Machine Intelligence Institute (Amii)
📍 Edmonton