Postdocs in Machine Learning for Olfaction

About the role

Multiple two- to three-year postdoctoral positions at KTH on machine learning for the sense of smell, predicting odour perception from molecular structure, learning olfactory embeddings, and building generative models for new fragrance molecules. The work draws on chemoinformatics, neuroscience, and modern deep learning. Deadline 18 May 2026.

Responsibilities

• Build and train neural models that map molecular structure to odour descriptors.
• Curate and analyse olfactory datasets, Pyrfume, GoodScents, and partner-provided industrial data.
• Publish in NeurIPS, ICML, JCIM, or related venues.
• Collaborate with experimental olfaction researchers and industrial partners.

Requirements

• Doctoral degree in machine learning, computer science, computational chemistry, neuroscience, or a related field, defended by the start date.
• Strong deep learning experience, including familiarity with graph neural networks or transformers on chemistry data.
• Python proficiency with PyTorch, RDKit, or similar.
• Fluent English.

Nice to have

• Prior work on chemoinformatics, drug discovery, or scent.
• Publications at top ML venues.
• Familiarity with olfactory psychophysics or sensory neuroscience.

How to apply

Search “Machine Learning for Olfaction” on KTH’s Varbi portal at the apply link. Submit CV, transcripts, research statement, and the “Five most meritorious scientific articles” document before 18 May 2026.