The position is under the supervision of Dr. Michal Koziarski within the Molecular Medicine program at The Hospital for Sick Children, University of Toronto, and the Vector Institute. Our research group focuses on developing machine learning-based pipelines that leverage generative models and active learning to efficiently explore chemical space, with applications in drug discovery. We are seeking a candidate interested in fundamental ML research driven by real-world applications.
The role will focus on developing generative models for small molecule design. The candidate will contribute to the development of chemistry-informed generative models using frameworks such as generative flow networks, reinforcement learning, diffusion, and flow matching, guided by practical considerations of high-throughput chemical synthesis. In addition to algorithm development, the candidate will have the opportunity to apply these methods in multidisciplinary projects bridging computational design and experimental validation in the wet lab.
What We Offer:
-
An interdisciplinary environment with expertise and close collaborations spanning core ML, chemistry, and biology.
-
ML research fundamentally motivated by the most pressing biological questions.
-
Flexibility in shaping research directions in alignment with the group's mission and the candidate's interests.
-
Access to dedicated computational resources at SickKids Research Institute and the Vector Institute cluster.
-
Engagement with the Vector Institute and Acceleration Consortium ecosystems.
Here's What You'll Need:
-
PhD in a machine learning-related discipline.
-
Strong publication track record in top ML, computational chemistry, or computational biology conferences and journals.
-
Proficiency in Python, demonstrated through completed projects, preferably with publicly available repositories.
-
Experience with deep learning frameworks such as PyTorch and JAX.
-
Hands-on experience in developing and training deep learning models.
-
Familiarity with reinforcement learning and/or generative models.
-
A commitment to understanding and aiding in the pursuit of equity, diversity and inclusion objectives.
Preferred Qualifications:
-
Experience in drug discovery and chemical data.
-
Background in generative models for small molecule design.
-
Experience with GFlowNets.
-
Familiarity with active learning approaches.
Employment Type: Full-time temporary 1 year contract
