Alumnus Liam Kaufman’s entrepreneurial path in digital health innovation

As the University of Toronto celebrates Entrepreneurship Week 2026 from March 2 to 6 — a showcase of innovation, startup success and bold ideas across the tri-campus community — we are highlighting alumni who embody that entrepreneurial spirit. Liam Kaufman is one such graduate, translating cutting-edge research into impactful health technologies and building ventures that bridge science and industry.

Across roles as an entrepreneur, scientist, engineer and strategic leader, Kaufman has built a career focused on translating advanced AI and clinical research into real‑world health care tools.

After completing his BSc in psychology at Western University, Kaufman earned a master’s degree in medical science at the University of Toronto’s Faculty of Medicine (now known as the Temerty Faculty of Medicine) in 2008 and a BSc in computer science in 2011, also from U of T.

Kaufman has always had an entrepreneurial spirit. As a child he went door-to-door shoveling neighbours’ driveways for money and even made crafts to sell at his father’s birthday party. His first adult success came shortly after graduating from U of T, with Understoodit — a tool for collecting anonymous feedback during class. The platform gained international media attention before being acquired by EventMobi.

Currently, Kaufman serves as executive vice president of product and academic at Cambridge Cognition, where he helps guide the company’s global strategy in cognitive assessment technologies and digital biomarkers. Before joining Cambridge Cognition, he was the co‑founder and CEO of Winterlight Labs, which develops speech‑based digital biomarkers for cognitive impairment and mental health (acquired by Cambridge Cognition in 2023).

We talked to Kaufman about his path to working at the intersection of neuroscience, machine learning and digital health innovation.
 

How did you become interested in neuroscience?

I did my undergrad at Western in psychology and kept gravitating to the science side —stats, methods, functional MRI. I’d also been reading pop‑neuroscience books and was captivated by how scientists use tools and methodology to explore how we think and learn. After graduating, I worked at BC Children’s Hospital as an MRI tech/research assistant, which let me apply what I’d learned in a real clinical setting. I liked the rigour and objectivity of science, and neuroscience felt like the intersection of what I loved — plus I wanted to work with patients and see what I was learning in action, day to day.

For your postgrad, how did you land on the Institute of Medical Science at U of T?

I wanted something applied, and IMS put me in a hospital environment (Sunnybrook Health Sciences Centre) working directly with patients, not just in a theoretical or purely academic context.

Candidly, the stipend also mattered. Toronto isn’t cheap for grad students, and IMS had one of the highest stipends, which helped.

The program catered to clinicians and residents, so I didn’t have to TA and could focus on research and data collection. Working with Sandra Black (MD ’78, PGME Neurology) was formative: high rigour, high expectations. I learned to only say what I could back with evidence and got a lot of practice presenting to committees, which was great for building confidence and learning how to talk with experts who know more than you.

What did you study?

My thesis focused on a specific eye‑movement task called the anti‑saccade task. Normally, when something appears in your peripheral vision, you automatically look toward it. We trained people to look in the opposite direction, which requires executive control to inhibit that automatic gaze. Healthy people are generally good at this, but when the frontal lobes are damaged, the task becomes much harder. Alzheimer’s is usually thought of as a memory disorder affecting the temporal lobes, but what we showed was that people with Alzheimer’s and mild cognitive impairment had clear difficulties with this task — they were much more likely to look toward the stimulus. I did a meta‑analysis and published our findings, adding more evidence that Alzheimer’s involves impairments beyond memory.

What prompted you to pivot to computer programming?

I planned to do a PhD and had strong support. But after a late night prepping for a talk, I asked myself if that’s what I wanted for the next three to four years — especially given how competitive hospital scientist jobs are. Meanwhile, I’d taught myself enough programming for side projects and data analyses to realize I liked the challenge and the tangible problem‑solving. Employment prospects also looked stronger, so I decided to bridge the two fields. I hadn’t taken math in years, so I blitzed grade‑10 through grade‑12 material in a few months to be adequately prepared for computer science at U of T. In retrospect, having both skill sets has been really useful.

How did you get your start as a digital health entrepreneur?

Right after graduating, I launched Understood.it. It got good press — CTV, Toronto Star, even the front page of TechCrunch — which gave me a taste of early traction. EventMobi acqui‑hired us; they were more interested in the team than the product, and I led their mobile app group as a developer/manager.

I still wanted to get back to the neuroscience-computer science intersection, so in 2015 I met with Frank Rudzicz who was a U of T faculty member at the University Health Network’s Toronto Rehabilitation Institute at the time. His expertise was in computational linguistics and natural language processing, and his research showed you could probably detect Alzheimer’s with about a minute of speech. I found the work intellectually captivating and I could see the potential for commercialization. I left my job, taught a computer science course to patch together income, and with two of Frank’s grad students we started Winterlight Labs that fall.

How has your medical science education at U of T helped you in your career?

Sandra’s mentorship taught me rigour: if I’m going to say something, I need evidence. As an entrepreneur, that translates directly to how I prepare for investors and customers —choosing words carefully, anticipating questions and backing up claims.

IMS also forced me into regular, polished presentations to advisory committees, which made me a better public speaker and more comfortable engaging experts.

Beyond training, the U of T ecosystem mattered. Winterlight went through Rotman’s Creative Destruction Lab and the Temerty Faculty of Medicine’s Health Innovation Hub (H2i). H2i made pivotal introductions that helped us get pharma traction and funding. U of T’s combination of strong medical research and strong AI created the right environment to build at that intersection.

How are you evolving your product and business now? What’s on the horizon?

The business exploded during COVID, but in 2023–2024 it was tough — biotech funding dropped and studies slowed. In 2025 we’ve seen a real rebound. The tech we’ve built over 10-plus years is now in a lot of trials. We started in Alzheimer’s and, since 2019, have been expanding into schizophrenia and depression. Pharma increasingly wants to measure what matters to patients — communication, memory, orientation — which aligns with our approach.

On the tech side, we’re adding languages (we support nine or 10 now and keep adding), automating more and scaling. Speech is captured in basically every central nervous system clinical trial for quality assurance, so there’s opportunity to analyze speech alongside third‑party assessments — and potentially in health care more broadly, analyzing doctor–patient conversations with consent. We’re still just scratching the surface.

— Original story by Temerty Faculty of Medicine

U of T establishes new Hinton Chair in Artificial Intelligence thanks to generous support from Google

Geoffrey Hinton at U of T’s St. George campus. Photo by Nick Iwanyshyn.

The University of Toronto is proud to announce that it has established the Hinton Chair in Artificial Intelligence, made possible by $10 million in funding from Google.

This new chair will honour the extraordinary legacy of University Professor Emeritus and Nobel laureate Geoffrey Hinton at U of T and Google by enabling the university to recruit and retain another brilliant, internationally recognized AI expert to make profound contributions to the field.

“On behalf of the university, I would like to express our deepest gratitude to Google for this wonderful investment,” said Melanie A. Woodin, University of Toronto president. “This new chair will enable us to build on Geoff Hinton’s historic contributions in artificial intelligence and to advance our record of transformational research in fields of crucial importance to the world.”

U of T is matching Google’s support with an additional $10 million in funding. This historic $20-million investment makes the Hinton Chair in Artificial Intelligence one of the University of Toronto’s most prestigious and generously supported advanced research roles, with substantial endowed support for a leading-edge AI researcher and additional funds to drive fundamental discoveries and insights — creating the intellectual underpinnings necessary to take AI to the next level.

“Google is proud to partner with the University of Toronto in establishing this endowed chair, recognizing the extraordinary impact of Geoff Hinton, whose Nobel Prize-winning work laid the foundation for modern artificial intelligence,” said Jeff Dean, chief scientist at Google DeepMind and Google Research. “On a personal level, it was a delight to have Geoff as a colleague for more than a decade. This chair will empower world-class academic scholars to accelerate breakthrough innovations and drive responsible research that shapes a future where AI serves a common good.”

The Hinton Chair is the first in the university’s newly developed Third-Century Chairs program, a strategic effort established on the cusp of U of T’s bicentennial to attract and retain visionary scholars who can transform disciplines, shape global discussions, improve lives and strengthen Canada’s capacity to prosper. With competition for talent at an all-time high, the program will help the university amass critical expertise in areas essential to the country’s future — a key priority shared by the Canadian government, which recently announced a $1.7-billion commitment to attract top global research talent.

The Hinton Chair will also help U of T recruit, teach and train some of the world’s most talented students in the field, fuelling innovation in AI applications across medicine, engineering, discovery science, the humanities and more, expanding the university’s AI networks and international partnerships and sparking a new wave of promising AI startups.

Building on Hinton’s revolutionary research

The Hinton Chair in Artificial Intelligence aims to support the same brilliant, exploratory research that its namesake has pursued during his time at the University of Toronto and at Google.

After receiving his PhD in Artificial Intelligence from the University of Edinburgh in 1978 and completing several years of postdoctoral work in the United Kingdom and the United States, Geoffrey Hinton came to the University of Toronto in 1987 as a fellow of the Canadian Institute for Advanced Research (CIFAR). There, along with several graduate students, he accelerated his expansive work on artificial neural networks as a potential pathway for advancing AI, developing core concepts such as: backpropagation algorithms; distributed representations; time-delay neural nets; mixtures of experts, variational learning and deep learning; and, most famously, Boltzmann machines.

In the 2000s, Hinton’s ideas began to yield extremely promising results. In March 2013, as more tech companies recognized the promise of artificial neural networks, Hinton joined Google as a vice president and engineering fellow, where he would stay for the next decade, splitting his time between the company and U of T.

Although many people have contributed to the current state of AI, arguably none was more important than Hinton, whose decades-long research forms the foundation of modern artificial intelligence and its myriad applications across nearly every discipline and sector. He is also responsible for the “Hinton effect,” which saw many of his students go on to lead AI advances in universities and companies across the globe.

“I am grateful for having been able to pursue my research at the University of Toronto, which afforded me the time and resources to develop the ideas that would eventually grow into the success of neural nets,” said Geoffrey Hinton. “I am encouraged that the Hinton Chair in Artificial Intelligence will support the next generation of AI research in the same vein, allowing ideas of great promise to germinate for the benefit of all humanity.”

Together with John J. Hopfield, Hinton won the Nobel Prize in Physics in 2024 for his foundational work in enabling deep learning and propelling the field to its current peak.

University of Toronto — a world leader in AI

Based at the Faculty of Arts & Science’s Department of Computer Science — ranked 12th in the world according to the 2025 QS World University Rankings by Subject and a global leader in deep learning and generative AI — the Hinton Chair in Artificial Intelligence will leverage U of T’s and Toronto’s substantial and widely recognized strengths in AI.

“It’s thrilling to consider the astonishing possibilities of welcoming a globally leading AI researcher into this setting,” said Interim Dean, Faculty of Arts & Science Stephen Wright. “At the Department of Computer Science, the chair-holder will be surrounded by a remarkable concentration of scientific knowledge and creative skills, and a deep, proven track record of research excellence. It’s an ideal platform for charting new pathways and pursuing breakthrough discoveries in our shared goal of a brighter technological future for all.”

The University of Toronto is home to CIFAR AI Chairs and Canada Research Chairs in AI and has spurred several cutting-edge AI startups such as BlueDot (infectious disease intelligence), Waabi (autonomous trucks) and Deep Genomics (RNA-focused AI for disease detection). In addition to Hinton’s Nobel Prize, U of T’s faculty members and graduates have earned many other distinctions, including two Turing Awards, two of the three Herzberg Gold Medals ever awarded to computer scientists, and 15 Sloan Research Fellowships.

The university also consistently attracts and trains the best and most diverse cohort of undergraduate and graduate students from around the world, with hundreds pursuing AI-related studies across the university.

​In addition, U of T is home to an array of AI-focused research initiatives such as the Acceleration Consortium, the Schwartz Reisman Institute for Technology and Society, the Data Sciences Institute and the Temerty Centre for AI Research and Education in Medicine. The university also maintains a close partnership with the Vector Institute, a globally renowned organization co-founded by Geoffrey Hinton that empowers researchers, businesses and governments to develop and adopt AI responsibly.

An impactful partnership: Google and U of T

Establishing the Hinton Chair in Artificial Intelligence is the latest instance of U of T and Google’s longtime partnership in supporting discovery-based research. Over the years, Google has engaged many AI-focused U of T alumni and academic leaders, including Hinton, and the two organizations are founding partners in Toronto’s Vector Institute. Previous funding from Google has helped position the University of Toronto as a preeminent centre for advanced research in AI, and this new chair will greatly expand this impact.

“We are extremely grateful to Google for partnering with us to establish a chair dedicated to cutting-edge research on the defining technology of our time, which will help generate societal and economic benefits for communities across the planet,” said David Palmer, U of T vice-president, advancement. “Hinton himself once said that real breakthroughs come from people focusing on what they’re excited about, and the Hinton Chair will honour this example by providing unprecedented support for the next era of elemental, curiosity-driven work in artificial intelligence.”

— Original story by the University of Toronto

A&S alumni mentor students and recent grads at latest backpack 2 Briefcase industry night

Computer Science alumna Julie Chan delivers her keynote address at the latest b2B career night.
(Photo credit: Bilal Khan)

Arts & Science students and recent graduates met alumni mentors for an evening of casual conversation and networking at the latest backpack 2 Briefcase (b2B) industry night — and they came away with great advice.

“As you move from backpack to briefcase, the one thing that can really set you apart is being willing to step outside, meet people and connect in real life,” says Julie Chan, the keynote speaker and Department of Computer Science alumna.

The b2B program connects A&S alumni with students to help them make the transition from university to a rewarding career. Industry nights include a keynote address and casual conversations that give students the opportunity to explore their next career steps.

Chan, who earned her bachelor of science degree in 1982 as a member of New College, has made mentoring a top priority throughout her career and stayed connected to the Department of Computer Science. At b2B, she shared a story about coaching a mentee through a job interview, which led him to land a role at his dream company.

“Mentors are a gateway to wider networks; they will know other experts who can help you,” Chan says. “Don’t hesitate to ask for introductions as I did when I was looking for work.”

The keynote address resonated with attendees of the event.

“I really liked Julie’s advice about stepping out and meeting people in person,” says Christina Sun, a second-year studying political science, sociology and environmental studies as a member of Woodsworth College. “It’s good to build those human connections.”

Department of Computer Science announces promotion of five faculty members

The Department of Computer Science is pleased to announce the promotions of the following faculty members, effective July 1, 2025:

• Murat Erdogdu promoted to the rank of Associate Professor with tenure

• Fan Long promoted to the rank of Associate Professor with tenure

• Kirill Serkh promoted to the rank of Associate Professor with tenure

• Bo Wang promoted to the rank of Associate Professor with tenure

• Tovi Grossman promoted to the rank of full Professor

University of Toronto team discovers vulnerability at hardware-software boundary in cloud systems

From left to right: David Lie, director of the Schwartz Reisman Institute, Gururaj Saileshwar, assistant professor in the Department of Computer Science, and Yuqin Yan, a student at the Department of Electrical & Computer Engineering, discovered a security flaw in AMD’s cloud protection technology, revealing how interactions between hardware and software can expose sensitive data. (Photos: provided)

Cloud computing has become an essential part of our everyday lives, both personally and professionally. Whether it’s storing family photos, running a business or training cutting-edge AI models, we rely on remote servers to keep our data safe and secure and trust that it won’t be modified in any way.

Although storing information in the cloud exposes data to potential risks, hardware vendors like AMD mitigate these risks by collaborating with major cloud providers such as Amazon Web Services (AWS), Google Cloud and Microsoft Azure, to provide hardware-level protection that is meant to keep data secure and confidential even if the cloud provider experiences a security breach.

However, a team of University of Toronto researchers led by David Lie, director of the Schwartz Reisman Institute (SRI) and Gururaj Saileshwar, assistant professor in the Department of Computer Science, and executed by Yuqin Yan, a student at the Department of Electrical & Computer Engineering (ECE), found a flaw in these systems. They discovered that the complex interactions between the software that the cloud providers run, and the hardware-level protection, leads to new security challenges and vulnerabilities.

“Unlike most security vulnerabilities that are found in either the hardware or the software, what sets this discovery apart is that it was found in the interplay between the software and AMD’s hardware” said Lie, who is cross-appointed to the Department of Computer Science. “In this case, it was found when the hypervisor and central processing unit (CPU) interacted.”

We can think of a hypervisor as the “virtual landlord” of AMD’s chips. It is software that “rents” out computing resources, such as memory, to the cloud customer “tenants” allowing various customer workloads to run securely, independently and confidentially on its CPU.

AMD’s confidential computing technology is designed to protect such tenants in the event that the landlord is controlled by a malicious entity; in other words, if it is hacked. It encrypts data in a way that depends on its location within memory, so if the same data is stored in two places, it is encrypted completely differently. That makes it difficult for the hypervisor to know anything about the data or track it across locations, increasing the security of the data.

“The system lets the hypervisor move data around to manage memory efficiently,” explained Lie. “So when data is relocated, AMD’s hardware decrypts it from the old location and re-encrypts it for the new location. But, what we found was that by doing this over and over again, a malicious hypervisor can learn recurring patterns from within the data, which could lead to privacy breaches.”

Vulnerabilities like this have the potential to affect people and organizations alike.

“These are the kinds of unexpected consequences that come from the complexity of modern systems,” said Saileshwar. “The attack we discovered, which we call Relocate-Vote, shows how that complexity, especially at the boundary between secure hardware and untrusted software, can lead to serious vulnerabilities.”

The majority of the research was performed by ECE student Yuqin Yan. It also included now-graduated ECE student Wei Huang, ECE and SRI Postdoctoral Fellow Ilya Grishchenko, and UBC faculty member Aastha Mehta.

“Our role in academia is to identify vulnerabilities in real systems,” said Saileshwar. “I am proud of the work our team did. We are pleased that Yuqin was able to present this paper at the USENIX Security Symposium in Seattle, Washington.”

Going forward, Saileshwar notes that the consequences of hardware security are only going to grow and affect more organizations over time.

“As we move more of our data to the cloud, hardware security is becoming more important than ever,” said Saileshwar. “Hardware is becoming more complex, it’s adding more features all the time, and we’re relying on its security features even more. We’re placing a lot of trust in hardware, making the research our team is doing at the University of Toronto into hardware security issues more impactful than ever.”

For more information about Relocate-Vote, please visit the project website.

Original story by Andrea Wiseman for the Schwartz Reisman Institute