Postdoctoral Researcher NLP for Evidence (GALENOS Project)

Postdoctoral Researcher NLP for Evidence (GALENOS Project)

Idiap Research Institute

Martigny, Switzerland

Role Summary

The Human-Centred Health AI research group develops and applies artificial intelligence approaches to advance health, spanning the molecular, individual and societal scales. Our research has three main focus areas: AI for Biomolecules, AI for Evidence, and AI for Humans in Society.

We are looking for a talented and motivated Postdoctoral Researcher to join our interdisciplinary team working at the intersection of computational methods and health science, for the Wellcome Trust-funded GALENOS project. You will develop, improve, and apply state-of-the-art natural language processing (NLP) and time series modelling techniques to the published scientific literature on mental health, with the goal of accelerating evidence synthesis, identifying trends in research, and generating novel insights that can inform policy and practice. This is an exciting opportunity to do methodologically rigorous work with real-world impact.

Your specific tasks will include training and fine-tuning language models and time-series models, analysing research data, writing research publications, collaborating with a global interdisciplinary team, and contributing to the development of public-facing web interfaces for users to interact with models.

Your Profile

Must have:

• PhD (or equivalent research experience) in a relevant field such as computer science, computational linguistics, data science, statistics, or a quantitative social or health science
• Strong programming skills in Python
• Experience with time series analysis as well as machine learning frameworks such as PyTorch and Transformers
• A track record of conducting rigorous, reproducible research, evidenced by publications and project outputs
• Strong written and verbal communication skills in English

Nice to have:

• Experience working with biomedical or scientific literature, including PubMed, MEDLINE, or similar corpora
• Familiarity with evidence synthesis methodologies such as systematic reviews or meta-analysis
• Experience fine-tuning large language models for domain-specific tasks
• Familiarity with web development would be an asset
• Contributions to open-source projects or publicly available research code would be an asset
• Familiarity with the mental health research domain would be an asset