Sean Caffrey | Executive Director of Acceleration Consortium
University of Toronto

Sean Caffrey, Executive Director of Acceleration Consortium, University of Toronto

Sean Caffrey is the Executive Director of the Acceleration Consortium, a global research initiative led by the University of Toronto (UofT) that is transforming scientific discovery. The Acceleration Consortium joins materials science with artificial intelligence, robotics and advanced computing to build self-driving labs that can discover new materials 100 times faster. Rapid materials discovery is critical to addressing society's global challenges. Sean helped to launch the Acceleration Consortium while working at the office of the Vice-President of Research and Innovation as the Executive Director, Strategic Initiatives Development. In this role, Sean led a team developing and scaling high-priority interdisciplinary research networks and supporting these networks through fundraising and partnership development with industry, public sector research institutes, and not-for-profit agencies. Before joining the UofT, Sean worked in several capacities at Genome Alberta.

Sean received an MBA specializing in entrepreneurship and innovation, a Ph.D. in Microbial Genomics, and a Project Management Professional certification. In addition to his interest in research, Sean is a professional photographer, and his travel photography has appeared in major publications such as National Geographic


Day 1 (26th June) @ 14:15

The lab of the future is self-driving

Chemical space includes an incredibly large number of potential molecules, exceeding 1060. This space is far too large for scientists to enumerate, let alone evaluate. AI-driven autonomous labs (self-driving labs or SDLs) are needed to optimize the exploration of diverse material compositions to discover novel materials efficiently. Even when the exploration space is constrained, SDLs can improve the reproducibility of results and reduce the number of experiments and resources required, thereby reducing the cost and time required for discovery.

Although SDLs have been used to study classes of materials with high-dimensional spaces, such as high-entropy metals, many challenges must be addressed to produce generalizable, scalable SDLs. These range from small material datasets for training machine learning systems to the difficulty of synthesizing materials with complex reactions and steps and autonomy testing a wide range of material properties.  

This talk will discuss how the Acceleration Consortium is addressing these challenges and provide examples of how SDLs have deployed to dramatically accelerate materials discovery.

last published: 14/Jun/24 11:25 GMT

back to speakers



For conference production and speaking opportunites:

Für die Produktion von Konferenzen und die Möglichkeit, Vorträge zu halten: 

For sponsorship and exhibition opportunities: