Hanan Ather

I am a Member of Technical Staff at the Centre for AI Research and Excellence (CAIRE) at Statistics Canada. I build and deploy AI systems for federal stakeholders, with a focus on evaluation, safety, and security across large language models and agentic systems.

I also develop technical guidance for governing and securely deploying these systems in settings that handle confidential data.

As a member of the Office of Responsible AI, I lead technical reviews of AI systems across government and contribute to frameworks and governance guidance for UNECE, G7 GovAI, and the International Statistical Institute (ISI).

• First author of a paper on governing LLM services that handle confidential data, published in Journal of Official Statistics.

• Won first place in the 2025 IAOS Young Statisticians Prize for LLM-assisted record linkage, published in Statistical Journal of the IAOS.

• Contributed to the development of a United Nations handbook, Uncertainty Quantification in ML-Enhanced Statistical Inference.

• Master of Science in Mathematics and Statistics at the University of Ottawa

  • Research topic: Deep Reinforcement Learning and Function Optimization

Projects

• I built Probability Lab, an interactive probability and statistics platform currently in use at the University of Ottawa’s second-year engineering program.
• I published a 6-part series of reinforcement learning labs (Jupyter notebooks): Reinforcement Learning Labs.

Graduate Course Notes

• Probability Theory II
• Real Analysis
• Statistical Machine Learning
• Graduate Probability Theory
• Mathematical Statistics II
• Mathematical Statistics I

Papers & Publications

• Governing Large Language Model Services for Confidential Microdata: A Decision Framework for National Statistical Offices Hanan Ather, Mohammed Haddou, and Saeid Molladavoudi. Journal of Official Statistics, 2026.
• LLM-assisted record linkage: A framework for official statistics Hanan Ather. Statistical Journal of the IAOS, 2026.
• Uncertainty Quantification in ML-Enhanced Statistical Inference — A Practitioner Handbook with Applications in Official Statistics UNECE · 2026 · Contributor
• Master Thesis: Deep Reinforcement Learning
• Law of the Iterated Logarithm
• MapReduce for Parallel Machine Learning

2026 Award of Excellence in Research

Selection Committee, Modern Statistical Methods and Data Science Branch, Statistics Canada

“Hanan has established himself as an emerging expert in applying statistical methodology, machine learning and LLMs to official statistics. He received the 2025 IAOS Prize for Young Statisticians for his paper on LLM-assisted record linkage, presented at the 65th ISI World Statistics Congress. He also co-authored a decision framework for governing LLM services involving confidential microdata and contributed to international guidance on uncertainty quantification in ML-enhanced inference. His research advances the statistical foundations of ML-assisted official statistics, including prediction-powered inference, conformal prediction, and uncertainty quantification for LLM-assisted coding and record linkage. Hanan also advances research on responsible agentic AI adoption.”

2025 IAOS Young Statisticians Prize (First place)

International Association for Official Statistics