3 Archetypes of Canadian AI Founders
Alex McIsaac here! (X)
In a year dominated by the meteoric rise of AI we observed the emergence of three distinct archetypes of Canadian founders building AI startups.
Our insights are derived from the early-stage sourcing activities at Northside Ventures where we focus on investing in Canadian founders throughout the United States and Canada. We engage at the very inception of company formation and see a high volume of deal flow, sourcing between 3,000 and 5,000 companies annually.
These three archetypes help illustrate our approach to the categorization of opportunities and have been shaped by two structural trends we’ve observed across the Canadian venture ecosystem:
- Two Canadian universities stand out from the rest when looking at the founders of AI companies, University of Toronto and the University of Waterloo, that serve not only as centers of education but also as focal points for pioneering developments in AI and are recognized for the talent they produce on a global scale.
- The post-COVID surge in remote work along with the growing propensity of US funds to invest beyond conventional tech centers are reshaping the landscape and while this trend is more than a decade in motion it has accelerated in recent years. This shift enables founders to cross borders with greater ease, diminishing constraints in hiring top talent and accessing capital.
Collectively these observations influence the three archetypes that together, underscore the versatility and impact of Canadian AI talent and position Canada as a significant player in the global AI industry.
Archetype #1: Founders educated in Canada move to San Francisco to build (ie. OpenAI, Databricks)
This archetype is characterized by AI researchers trained under Geoffrey Hinton or Yoshua Bengio; two famed Canadian AI leaders globally recognized as the top minds in AI who have proven to be talent creation machines. We’ve observed many of their researchers move to larger markets, such as the Bay Area, where arguably the highest number of talented AI engineers have aggregated.
The most notable example is OpenAI and its co-founder and Chief Scientist, Ilya Sutskever who was a student in the Machine Learning group at University of Toronto, working with Geoffrey Hinton, before working as a postdoc at Stanford in Andrew Ng’s group and moving on to Google Brain. Andrej Karpathy was also a founding member of OpenAI and first got into deep learning by attending Hinton’s class at University of Toronto, while also working in Fei-Fei Li’s Vision Lab at Stanford and holding a role as Sr. Director of AI at Tesla. We noticed Open AI has a strong contingent of Canadian employees; according to LinkedIn data there are 60 employees (8% of all OpenAI employees) who attended a Canadian University including University of Waterloo (17), University of Toronto (8), Queen’s University (8) and McGill (4) and University of British Columbia (UBC) (4).
Another example is Databricks, an AI infrastructure B2B software company recently valued at $43B, that was co-founded by University of Waterloo and University of Toronto graduates, Matei Zaharia and Reynold Xin (amongst several other co-founders). Around 2009 both Matei and Reynold moved to California to work at UC Berkeley and Stanford and created Apache Spark where they teamed up with the other Databricks co-founders and founded Databricks in 2013.
We compared the number of employees that graduated from a Canadian university across a few of the top AI companies in the US and found a strong presence of Canadian graduates:
Here we see two Canadian universities stand out from the rest — the University of Waterloo (UW) unanimously has the greatest presence across the top AI companies followed by the University of Toronto (UofT). UBC is third with McGill, Queen’s and Université de Montreal making an appearance. It’s clear to us these universities are generating top-tier graduates in computer and data science, who possess the skills to land roles at leading AI firms in the US.
While many Canadians see the departure of talent as negative, there is an argument that these Canadian university graduates are accumulating valuable experience scaling some of the largest AI companies in the world. We anticipate many will generate new ideas and venture out to establish their own companies, potentially ushering in a surge of exciting AI startups led by Canadian entrepreneurs.
Archetype #2: Founders educated in Canada who stay here to build (ie. Cohere, Ideogram)
This founder profile is characterized by a Canadian, often an immigrant to Canada, who was educated at a Canadian school such as University of Toronto or University of Waterloo, and often spent time working under Hinton or Bengio.
A prime example is Cohere, a provider of Large Language Models (LLMs) and a recent Canadian unicorn, founded by University of Toronto graduates Aidan Gomez, Nick Frosst and Ivan Zhang in 2019 after Aidan and Nick left Google Brain. The company recently raised a $270m Series C at a $2.1B valuation led by Inovia with participation from Salesforce Ventures, Nvidia, Oracle, Thomvest, Index Ventures and others. Cohere’s technological progress and achievement in raising capital epitomizes the potential of Toronto-based AI companies who have decided to build their headquarters in Toronto and have achieved global recognition and impact.
Another earlier stage example is Ideogram AI, a Toronto-based startup offering generative text-to-image AI tools, who raised a $16.5m seed round led by Andreessen Horowitz and Index Ventures. Ideogram’s CEO and co-founder Mohammad Norouzi completed his PhD in Computer Science at University of Toronto before spending nearly 7 years as a Senior Staff researcher at Google. The other cofounders, Chitwan Saharia, William Chan, and Jonathan Ho, spent time at OpenAI, Berkeley, Indian Institute of Technology in Bombay, MILA in Montreal, and Google Brain in Toronto.
This year we saw several other notable examples of Canadian AI companies raise Series A and B rounds from top investors including Arteria AI ($30m Series B led by GGV Capital), Odaia ($25m Series B led by Threshold Ventures), Spellbook ($11m Series A led by iNovia and Thomson Reuters) and CentML ($27m seed round led by Gradient Ventures and NVIDIA).
The Toronto-Waterloo corridor stands out as a hotbed of AI talent resulting in more local founders choosing to stay local knowing they can access their existing networks to hire world class talent. Helping to supply this talent pool are the AI institutes and accelerators like Creative Destruction Lab (CDL), the Vector Institute and University of Toronto Early Stage Technology (UTEST) Program that continue to function as a training ground and contributing factor to a founder’s decision on where to build.
We anticipate this archetype will become more common as the talent pool continues to grow and employees from companies like Cohere, leave to start their own companies.
Archetype #3: Founders educated in Canada move to the US for work experience but move back to Canada to build
This year we observed several highly experienced Canadians with experience working for big tech companies or unicorns in the Bay Area exit or leave these long-standing senior positions and move back to Canada to join existing growth stage companies or launch new companies.
These founders are characterized as having 10+ years’ experience and are often looking to start families or move close to extended family. The reasons for remaining in the Bay Area, which typically include access to talent and venture capital, are now accessible in a remote work format and these founders are able to maintain their network for hiring and capital irrespective of where they’re located.
Two examples of this are Cohere’s hiring of Martin Kon as President & COO, who was previously CFO of Google’s YouTube division, and moved from San Francisco to Toronto last year. And earlier this year we saw Waterloo-based Vidyard hired Jonathan Lister as COO where he was previously VP, Global Sales Solutions at LinkedIn, also moving from San Francisco to Toronto. In addition to these examples, we’ve seen early employees from scaled companies such as Instacart, Blend, Ripple and Square, leave senior positions and move back to Canada to start something new here.
The flow of talent back to Canada was initially spurred by the Trump administration’s policies in 2016–17 and subsequently accelerated by the COVID pandemic in 2020 that liberated individuals from their geographical constraints and resulted in many choosing to relocate back to Canada to be nearer to family, all while retaining their employment.
The other primary reason for top talent moving back to Canada worth mentioning has been the US H1B Visa used by immigrants to the US that has a 6 year limit. While the visa can be extended, based on our conversations with founders, it appears many are opting to move back to Canada given its open immigration policy and proximity to the US.
The upcoming 2024 Presidential Election result could potentially produce another influx of talent back to Canada.
Conclusion
This is a story of Canadian dynamism, where globally recognized academic institutions like the University of Toronto and the University of Waterloo are not just educational hubs but crucibles of AI innovation.
The blend of academic rigor and entrepreneurial ambition is accelerating Canadian AI founders onto the world stage, both domestically and internationally. Driven by the structural shift in remote work norms and the expanding investment reach of top international venture funds post-COVID, this movement highlights Canada’s growing prominence as a vibrant and influential player in the AI sector, underpinned by its unique blend of innovation, adaptability, and global outreach.
We are certain more world changing AI companies will be built by Canadian founders and we are excited to have a front row seat to this exceptional network.
