In this episode of Tank Talks, Matt Cohen sits down with Ali Asaria, Co-Founder of Transformer Lab, to unpack the less visible side of the AI boom, from broken machine learning tools to the rise of autonomous research agents. Ali shares what it really looks like inside modern AI development and why the biggest opportunity isn’t just using models, but having the ability to train, control, and improve them.

Ali also reflects on his journey building across multiple tech waves, from creating BrickBreaker on BlackBerry to scaling Well.ca and Tulip, and now tackling AI infrastructure with Transformer Lab. He breaks down the realities most founders don’t talk about, why great products lose deals, how long enterprise sales actually take, and why success often comes down to trust, timing, and people more than technology.

Beyond AI, the conversation takes a broader turn into the future of innovation. Ali challenges the tech industry, especially in Canada, to think bigger, rebuild public trust, and focus on solving real-world problems through ambitious “mega projects.” If you’re trying to separate AI hype from reality and understand where the real leverage is being created, this episode gives you a much clearer lens.

Building BrickBreaker on 150M Devices (00:02:41)

How a side project at BlackBerry turned into a global phenomenon. The early lesson that distribution beats perfection. Ali shares how building something simple but widely adopted gave him an early taste of scale. It also shaped his belief that getting into users’ hands fast matters more than polishing endlessly in isolation.

The Early Days of E-Commerce in Canada (00:05:36)

Packing boxes manually, hacking payment systems, and why investors believed e-commerce would never work in Canada. From manually processing credit cards to building infrastructure from scratch, Ali walks through how scrappy the early days really were. It’s a reminder that many “obvious” markets today once looked completely unworkable.

Scaling Well.ca and the McKesson Exit (00:08:18)

How relationships with partners turned into acquisition opportunities. The messy reality behind “successful exits.” Ali explains how long-term partnerships quietly set the stage for acquisition, even before it was intentional. He also highlights how unpredictable and fragile deals can be, even when they seem done.

Enterprise Sales Lessons from Tulip (00:11:19)

Why great products don’t win deals. Trust, relationships, and the human side of multi-million dollar contracts. Ali breaks down how enterprise sales are less about features and more about credibility and relationships built over time. He also shares how incumbents win not because they’re better, but because they’re already embedded.

The Hard Truth About Startup Life (00:13:52)

“90% hell, 10% fun.” What founders don’t talk about publicly and how to choose the right investors. Behind the highlight reels, Ali emphasizes how difficult the journey really is and how rarely things go to plan. Choosing the right partners becomes critical when things inevitably get hard.

The Moment AI Changed Everything (00:16:22)

Why language models shattered the belief that human intelligence couldn’t be replicated. Ali describes the exact moment his worldview shifted after seeing what LLMs could do. What once felt impossible suddenly became inevitable, changing how he thought about both technology and opportunity.

What Transformer Lab Actually Does (00:20:11)

Simplifying AI model training, orchestration, and infrastructure across local machines and massive GPU clusters. Ali explains how fragmented and complex current AI workflows are, especially for researchers. Transformer Lab aims to remove that friction and make building models far more accessible and efficient.

Scaling AI From One Machine to Thousands (00:23:14)

The technical leap required to move from hobbyist experimentation to full-scale AI labs. Moving from a single machine to distributed systems introduces massive complexity most developers never see. Ali breaks down why solving this unlock is essential for the next generation of AI builders.

AI Hype vs Reality (00:25:41)

Why Ali believes we may already have AGI, and why valuations still don’t make sense. Ali challenges the common narrative by arguing we’re closer to AGI than people admit. At the same time, he questions whether the current market can realistically justify the valuations we’re seeing.

Canada’s Startup Ecosystem: Challenges & Advantages (00:32:11)

Why geography matters less than mindset, and why building is always hard everywhere. Ali pushes back on the idea that location is the primary constraint for founders. Instead, he argues that resilience and ambition matter far more than where you’re building from.

Why Tech Has Lost Public Trust (00:34:12)

From rebels to power players, and what founders must do to rebuild credibility. Ali reflects on how the tech industry’s image has shifted over time and why that matters. Rebuilding trust requires focusing on real impact, not just growth or financial wins.

The Case for Mega Projects (00:38:09)

Why Canada needs bold, visible innovation bets that actually improve everyday life. Ali argues that large-scale, collaborative initiatives could realign public perception and drive meaningful progress. The key is solving problems people actually feel in their daily lives.

The Future of AI and Talent Sovereignty (00:41:28)

Why owning talent matters more than owning infrastructure in the AI race. Ali emphasizes that long-term advantage comes from people, not just technology or compute. Countries that develop and retain top talent will ultimately shape the future of AI.

About Ali Asaria

Ali Asaria is a serial entrepreneur and one of Canada’s most accomplished technology founders. He created the iconic BrickBreaker game on BlackBerry, founded Well.ca (later acquired by McKesson), and built Tulip into a leading enterprise retail platform backed by top-tier investors.

He is now the co-founder of Transformer Lab, an open-source platform designed to simplify and scale AI model development. His work focuses on democratizing access to AI infrastructure, enabling developers and organizations to build advanced models without the complexity traditionally required.

Ali is known for his bold thinking on AI, startup ecosystems, and the future of technology, often challenging conventional narratives around innovation and scale.

Connect with Ali Asaria on LinkedIn: https://www.linkedin.com/in/aliasaria/

Visit the Transformer Lab website: https://lab.cloud/

Connect with Matt Cohen on LinkedIn: https://ca.linkedin.com/in/matt-cohen1

Visit the Ripple Ventures website: https://www.rippleventures.com/

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit tanktalks.substack.com

Dead as a Dodo? Maybe Not! Colossal's Beth Shapiro on the Science of De-Extinction — and Moonshots

Future Around & Find Out

Advice for Moonshot Builders

The AI-powered Podcast Player