ML Research Assistant: Uncertainty & Data-Driven Design

McGill University; Montreal; On-site; GBP 20,000 - 28,000; Part time

Job summary

McGill University in Burnside Hall is seeking a part-time researcher to develop methods for uncertainty quantification and to contribute to data-driven experiments. You will design active-learning strategies, build Python pipelines for modelling, and engage in publications and grant reporting.

The role emphasizes ML, probabilistic modelling, materials-informed optimization, and reproducible research. English-language university environment; start date 2026-10-13 with a 2027-02-26 end date.

Qualifications

• PhD in a related field is an asset.
• Knowledge of ML, probabilistic modelling, uncertainty quantification, and interpretability.
• Experience in ML-guided materials design or optimization with publications.
• Proficiency in Python and reproducible research tools.

Responsibilities

• Develop methods for uncertainty quantification and calibration to ensure reliable predictions and risk control.
• Oversee and assist with the design of active-learning strategies and experiments to improve data efficiency and guide experimental decisions.
• Develop and maintain Python pipelines for statistical modelling, experimentation, and analysis.
• Contribute to scientific publications, conference presentations, and grant reporting.
• Develop verifiable retrieval and memory methods for scientific large language model (LLM) agents.

Skills

Machine learning knowledge

Probabilistic modelling

Uncertainty quantification

Model interpretability

Python proficiency

Education

PhD in Statistics, EE, Materials Science, ME, or related

Job description

McGill University in Burnside Hall is seeking a part-time researcher to develop methods for uncertainty quantification and to contribute to data-driven experiments. You will design active-learning strategies, build Python pipelines for modelling, and engage in publications and grant reporting.

The role emphasizes ML, probabilistic modelling, materials-informed optimization, and reproducible research. English-language university environment; start date 2026-10-13 with a 2027-02-26 end date.