Booming: An Inside Look at Seattle's AI Startup Scene

By: Nate Bek

Seattle’s AI startup boom isn’t loud, but it’s reshaping the city’s future in real time. 

Startups are multiplying, talent is clustering, and investors are taking notice. “We are, by any measure, the second-most important city in the United States for AI,” says Kirby Winfield, founding general partner at Ascend. 

While Silicon Valley still dominates, Seattle has quietly built an AI battleground — powered by a deep well of technical talent, the cloud giants that fuel modern AI, and world-class research institutions.

On KUOW’s Booming podcast, Joshua McNichols sits down with Kirby to break down the forces driving this surge, the hidden network of startups shaping the industry, and whether Seattle’s AI moment is built to last.

(Join us and consider donating to support local journalism.)

We’ve lightly edited this transcript for brevity, added references, and clarified a fact. Enjoy! — Nate 👾

---

Joshua McNichols: There's an invisible startup boom happening in Seattle right now. 

Kirby Winfield: I'd say, gun to my head, how many startups are operating today in Seattle with AI at the core of their technology? 1000, at least.

Tons of artificial intelligence companies have cropped up in the area, and they're rivaling sectors like health tech and computer software, but the AI startup scene looks a little different than other industries… Seattle's secret AI startup boom could transform our economy. But are we in a bubble, or is it here to stay? I sat down with an insider to find out.

His name is Kirby Winfield, and he's a venture capitalist for this company called Ascend. I started out by asking him to explain to me exactly what he does. 

I take money from investors, and I give that money to startup founders to help them build the future. I think the best analogy I've been able to make for folks who don't know about venture capital is that we're sort of like betters at a race track. There are lots of horses and lots of jockeys, and there are lots of ways to try to handicap the race. You might look at the horse's training speeds, or you might look at the horse's provenance, but in venture capital, it's much more about the jockey than the horse. So my job is to find the right jockey, the right startup founder, and then they take that technology horse and ride it to victory. That's the idea.

And when you're looking at jockeys, you're focusing on AI?

Our fund is entirely focused on artificial intelligence applied to enterprise problems. So basically, businesses selling to businesses technology solutions using AI.

So you're just the guy to talk to. 

I hope so. 

Okay, so can you rank Seattle's place in the AI world? I mean, where does it sit in relation to other cities?

So we typically say that Seattle is HQ2, or the second headquarters of the US for AI. Obviously, Silicon Valley is HQ1. There's just such a density of technology there. But we are, by any measure, the second most important city in the United States for AI. We have one in four of the AI software engineers in the US living in Seattle. We have the second largest population of software engineers in the US period. And about 65% of AI compute usage runs through the cloud. The three largest providers of which are based in Seattle.

But the cloud can be anywhere.

But the talent that runs the cloud and builds the software that allows the cloud to work for AI lives in Seattle.

OpenAI has opened an office here. We've got the Allen Institute for Artificial Intelligence. We've got Microsoft. These are companies I've heard of, but I've also heard that there's a hidden startup scene here. Can you give a couple examples of AI startups rooted in and around Seattle that could really change people's lives?

Absolutely. Probably the most popular one is read.ai. And this is just a comprehensive AI assistant that sits in your email inbox and in your Word documents and on your Zoom calls and coordinates information for you, so that you're always aware of the context of what you're doing and why you're doing it and what may have been related to it before. And this can work for personal situations as well. It keeps track of conversations and historical data that's associated with a given relationship. So this can be useful for, let's say, you're like me, and you have a child with a disability, and you're coordinating across dozens of service providers and caregivers and schools and doctors. That can be challenging to manage on your own and read.ai has software that puts that all at your fingertips. That can be really life changing for people. 

That one speaks to me because it's so complicated even to do a search in my email inbox for the right correspondence. If I had like a little personal assistant tying all those threads together and bringing together other parts of my life outside of email, that could be useful.

Yeah, it really is like having almost a precognition of what the needs of a certain conversation or experience are going to be, and then populating your dashboard, if you will, with all the information you will need before you even know you need it. It's pretty magical.

I've heard that there's not a lot of investor money flowing to startups in Seattle. So is AI an exception to that?

I'd say AI is an exception to that in almost every market in the world right now, because there really is this kind of Cambrian explosion happening, and so there's been a lot of money coming into AI. It helps that we have Microsoft, Amazon, Google Cloud, Apple's HQ2 for technology. We have, I think, 200 now technical headquarters for large tech companies based here. What happens when you have that kind of technical talent base is that those people want to leave those big companies and start their own little companies, and that does attract investment capital. Not only from people like me and other firms in Seattle, but we did some research, and 90% of the investment money in technology companies in Seattle comes from outside of Seattle.

So why is Seattle a good incubator for AI? Are there reasons besides the labor pool?

I think there are. We have probably the greatest independent center for excellence for AI in the United States, if not the world, with Paul Allen's Institute for Artificial Intelligence. That is a one-two punch when combined with the Allen School for Computer Science at the University of Washington. 

We've backed a number of PhDs out of the UW to start their own companies, and those people have huge influence on hundreds and hundreds and hundreds of very talented PhDs and masters students and computer scientists who then go on and leave this market and go to do their masters at Stanford, or go to start a company in Silicon Valley. And when their mentors start a company here, they come back. And so all of these centers for excellence become magnets for talent. That's one really important aspect. 

Certainly the hyperscalers — which is tech slang for Amazon and Microsoft and Google — also create a real bed of support. It's not just about the talent pool. If you're a software engineer who's an expert in AI, you want to apply that talent to big problems. And some of the biggest problems are ones that are experienced by these big companies that are located in Seattle. And so the problem space, if you're experimenting with AI technology, is really rich and dense in Seattle in a way that it's not really anywhere else in the country, maybe outside of Silicon Valley.

We've seen here in Seattle in the past, when an industry is booming, the benefits extend far beyond that particular industry. What kind of impact could AI have on the economy in Seattle, beyond just the companies themselves and making somebody like you money?

I think AI is poised to have an impact that maybe we don't understand on Seattle. An example that I always use is my daughter, who has Down syndrome. She's 15, and she will not be able to drive a car. Seattle is notoriously not a great place for public transit. For her to be able to take an autonomous vehicle, where I don't have to worry about her safety with regards to a driver, that's life changing for her and it'll be life changing for a lot of other people in this market in particular. 

You think about drug discovery. You think about energy. You think about the impact on maritime such as autonomous maritime vehicles. We've been talking about ferries here for 25 years, and the lack of skilled labor, aging of the ferry workforce, and the lack of persistently serviceable vessels. There's no reason we shouldn't have autonomous ferries in the next 10 to 15 years. 

I think a lot of what's going to change in Seattle is going to change a lot of other places, too. But there are those areas where I think we're probably going to see a little bit more impact just because of how our geography is positioned

Quick side note: I know for public transit nerds, they probably would take issue with the idea that public transit is bad here. We got Light Rail opening all over the place, all that kind of stuff. But I take your point. It's a car-oriented city. That's its legacy that it's trying to overcome.

We don't have the time to debate this here. The ridership numbers might suggest otherwise.

A topic for another time… So China rattled the markets recently with news that it's developing new, cheaper AI models like DeepSeek, for example. How could competition from China and other places affect our startup scene here in Seattle?

My perspective on China and DeepSeek, in particular, which is a new large language model, is that it was very disruptive because it actually outperforms some of the big closed-source models, like Open AI and Anthropic. So when it comes to that, I actually get excited, because if these state-of-the-art open source, low cost to free models become ubiquitous, that actually means that there will be a Cambrian explosion of AI applications and agents developed by companies that could not otherwise have existed because it's cost prohibitive to conduct the training of these models for an individual company. There's actually this paradox, Jevon's paradox, when the resource becomes freely available, there's an initial decrease in the volume of use, but then it should exponentially explode, just because it's so broadly available.

So DeepSeek and other cheaper, large language models could actually result in a startup explosion here, whether or not this is good or bad for Microsoft or other big AI companies here .You see this as being good for startups? 

Absolutely. 

One thing AI is making it easier to do is to start companies with very few employees, because you can use generative AI to do a lot of your engineering. So what does the low number of employees needed to start a company today mean for jobs in this region? Could we have an AI startup boom in Seattle that we don't really feel?

I think that's entirely possible. We're already seeing companies that we're backing hiring fewer people to do the same work. *Instagram, when it was bought by Facebook, had 13 employees — and it was bought for $1 billion. And that was before generative AI. So I absolutely think it's not a matter of if, but when, there's a billion-dollar company built by one person. 

Wow. 

That said, I do think one of the only things that's not going to be replaced are functions within a company that are inherently human. I was talking with someone today about the radical increase in AI-generated inbound sales emails. And there are all these startups that have raised hundreds of millions of dollars to build an AI sales representative. And what happens is, as a potential buyer, you realize it's all AI generated, and you delete everything. Before, if I got a cold sales email, I had to in the back of my mind, at least think, "Well, someone took the time to write this email and to notice something about my company or me that made them think that maybe I would buy their product," because they wouldn't do it if they didn't think I would, because their job is to sell their product. So there was this qualification that you could tell it happened by a human and it made you lean in a little bit and say, "Well, maybe not today." But if they email me again in six months or whatever, maybe I'll reply and say, "No, thank you," because I'm a nice person. And if I had their job, which I have, I would want someone to reply. That has been wiped off the face of the earth now. 

So actually what becomes valuable is warm, human introductions and personal networks. Which then begs the question of the return to office in this virtual world we live in and relationship building. And I think human relationships become the capital that's probably the most valuable in the next five years, at least, and I do worry a little bit about what that means for folks who are in the first decade of their career and may not have accumulated that may not know how to.

I've often thought that as AI progresses and becomes indistinguishable from humans, we will just move the bar — redefine what it means to be human, or we'll just keep moving the goal posts and cherish those things, or even fetishize those things that are sort of unique to human. Even if that means being chaotic.

I totally agree. 

We seem to go through these waves where everyone is investing in something, and then there's too much of it, and then the bubble pops. Can we just be in another one of those cycles?

We, without a doubt, are. There's nothing new under the sun, as Shakespeare said. I lived through the commercial Internet boom and bust, and operated startups through it. I lived through the mobile era of boom and bust and the cloud era. And now we're in the AI era, and just like those others, what you get is a very big rush of capital and excitement at the very beginning, when the technology is new. And the challenge is that that capital is chasing ideas that are not very well formed about what the future is going to look like. 

The future gets clearer. As it gets clearer, the costs for the technology tend to come down, which, combining that future being clearer and the cost coming down, results in a completely different landscape than what existed when the wave started to build. So that wave crests, and then the next wave comes. And I think we are just now at the cresting of the first wave, and the beginning of the next wave. And the next wave, I think, resembles the wave that came after the internet bust, where 80% plus of the commercial value on the internet today was created. And so that's what's exciting about investing in AI, at this moment in time is that I do think we're at the beginning of the real growth.

Stash Your Cash: Stifel Bank’s Minh Le on Startup Banking

By: Nate Bek

Startup founders flush with fresh cash face a whirlwind of priorities — milestones to hit, engineers to hire, and a product to build and polish. But a practical, often overlooked question looms: where should that cash live, and who can be trusted to manage it?

To get answers, we spoke with Minh Le, Managing Director at Stifel Bank and a veteran startup banker in Seattle. With two decades of tech banking experience, he has guided startups from their earliest funding rounds to successful IPOs. Along the way, he’s been a trusted advisor, securing lines of credit to help companies navigate turbulent times.

(Ascend banks with Stifel, part of a longterm relationship and trust of Minh.)

We’ve put together a detailed FAQ for founders seeking guidance on banking best practices. This resource offers practical insights for founders making financial decisions or seeking a trusted banking advisor.

*We've edited this conversation for brevity. Enjoy! — Nate 👾

---

Nate: Thanks for chatting with us, Minh. If I am a founder, how many banks should I have?

Minh: We recommend having two banks — one to serve as your primary bank and is a good strategic fit (has products and expertise that are tailored to your business), great service and can also go beyond banking to provide relevant business advice and connections to customers, talent, and investors; the other would service as a back-up that you have a couple of payrolls of cash at just in case your primary bank fails. A key component in this is making sure that your cash is also safe where you have it. 

Cash is critical for any company, so founders should know where its deposits are held, how secure the funds are, how accessible the funds are, and what they’re earning on the funds.

At Stifel, our clients benefit from our Insured Cash Sweep product, which allows them to keep their cash completely liquid (they can access their cash same day) and FDIC insured up to $250 million. The rate we pay is also very competitive for fully safe and fully liquid.

As VCs, we are constantly pitching our platform services and value add to founders. Do startup bankers offer similar support for founders? 

Different banks and bankers have different skill sets and approaches to working with their clients, some of which are mostly focused on their banking products and loans. At Stifel, we provide value to founders in two ways: focus and platform.

Focus: We largely work with founders and companies that are disruptive and often are venture backed (or are planning to raise venture capital), so our products and services are tailored to this profile. We have very experienced venture bankers that are focused on being great, strategic business partners for our clients. If a client needs banking services, we’re there to provide it, but often it’s other things like specific business advice or connections to investors, talent, customers, strategic partners, etc. that is a pressing need and that we can provide them. Also, the advice and the creative debt solutions are important as well. You can’t get those things from most banks.

Platform: We would argue we’re the best and provide the full platform, which includes all of those things mentioned above, but also extends to investment banking and wealth advisory (optimizing liquidity events for founders). Venture Bank + Investment Bank + Wealth Advisory can support a founder at all stages both personally and professionally, we believe that’s very compelling and unique to Stifel.

What are the most common banks startups are working with these days? 

Stifel Bank, Mercury, Brex, JP Morgan, Silicon Valley Bank

What are the pros/cons across various banks (big strategically important vs. regional)? 

You’ll get widely differing perspectives here depending on who you ask. I’ll try to be objective:

• Stifel – I’ll start with my naturally biased beliefs on Stifel, which I honestly believe:

• Pros:

  • Best Team - Our team has decades of experience supporting entrepreneurs. The person that ran the Tech practice nationally is with us as is the person that ran enterprise software nationally, the person that ran frontier tech national, the person that ran Corporate Banking nationally (the group that worked with late stage and public companies), I ran Washington and Western Canada, my counterpart that ran LA, etc.  

  • Broad Platform – Having the robustness of the expertise of our wealth advisory business and investment bank to pair with what we do on in the commercial bank for companies is a big differentiator, and quite frankly a vision we were trying to realize at SVB the last 5yr+ before the bank collapsed (we were building an investment and bought Boston Private to build out our private banking capabilities).  Stifel is the 7th largest wealth advisory business in the country by number of advisors, and half a trillion of assets under management and the leading middle market investment bank, very active in technology, healthcare, and related sectors, Stifel Investment Banking is #1 in equity deals under $1 billion market cap*. *Source: Dealogics M&A Analytics and 451 Research as of  8/19/24. Other company data available on www.stifel.com/investor-relations

  • Focused on Early to Growth - Having an affiliated investment bank focused on middle market companies is important as it shows Stifel has expertise more applicable to a broader swath of our clients. The larger banks can have a hard time getting the attention of their investment bankers that are incentivized to only work with the multi-billion dollar transactions, of which most tech outcomes are not.

  • The Right Size – We’re a $31B+ bank, which makes us large enough to work with large growth stage companies, while being very nimble in how we operate.  Larger banks (banks that are $200B+ like JPM, HSBC, and SVB) are considered LFIs(Large Financial Institutions) and have a substantially higher level of regulatory oversight that can limit what they can do and how they do things.

• Cons:

  • Limited Brand Awareness – We’ve come a long way since our west coast team was formed nearly 2 years ago (21 months to be exact).  When we lose to banks like JPM, it’s largely due to their brand and the feeling that they’re too big to fail.

  • UI/UX – We’re making huge strides here as well and are actually launching a new banking platform in 2025, but our UI/UX (nor will any other banks be) what folks experience with fintech players like Mercury. 

Other Pros and Cons:

• Pros:

  • Of the really big banks, HSBC, MUFG, and JPM have hired a lot of tech-focused bankers to try and capture the opportunity in tech.

  • HSBC, MUFG, and JPM are also viewed as too big to fail.

  • Mercury and Brex are software companies.  They tend to have the best UI and have digital account opening that tends to be the quickest.  They offer a good self-service offering.

• Cons:

  • Big banks tend to be more rigid and less nimble given their size and regulatory burden.

  • Mercury and Brex aren’t banks and can often mask the bank companies are actually working with. Underlying banks and their respective safety profile often aren’t evaluated, when they should be.  As an example, one of the leading banking as a service providers and has recently been in the news with meaningful regulatory issues that have emerged. Mercury and Brex are also both private, venture backed companies, so if you’re looking for transparent stability as a key factor in your bank, working directly with a regulated bank that’s public would be better at achieving that.

  • Fintechs are focused on UI and self-service. Those tend to be low value needs they can make good. They don’t have the relationship aspect of banking that drives real strategic value.

What about the newer entrants like Brex and Ramp if founders use them for other things?  

We actually partner with Ramp. They have a great credit card and expense management platform. Ramp and Brex are both really good at the credit card/expense management side of things. I think they should be leveraged for that. For banking, for reasons outlined above, I think choosing a stable bank to directly work with makes more sense, in particular one that understands your space and needs and can deliver value in those areas.

Addressing the elephant in the room, how can founders make sure they are FDIC-insured on all their cash? Or should they not worry about that? 

Founders should absolutely understand where their cash is and whether it’s liquid and safe.  FDIC insurance is one way to achieve safety of your cash. Insured Cash Sweep is a way to allow you to get FDIC insurance on potentially all of your cash. You can also have cash invested in treasuries as another option for safety.  

How can founders make sure they are getting decent interest on their cash? 

Taking a look at a few different options to compare is the way to ensure you’re getting a market rate. Also, look at what government treasuries are offering both short and near term as a good proxy. It’s worth nothing that Bank interest rates tend to move in line with interest rate changes made by the Federal Reserve. If the Federal Reserve decreases rates, you should expect the return on your cash to decrease as well. 

Should founders have business checking, savings, a sweep account, etc.? 

The most common way to do this is to have a checking account and sweep account. The sweep account will have some minimum peg balance and will sweep everything above that level. Just remember that what’s not swept is then typically subject to the $250,000 FDIC limit, so if the peg balance is $500,000, then $250,000 of that cash in there will not be FDIC insured. That's a small exposure, so maybe it’s okay. For instance, our peg balance is set at $250,000 and then the rest is swept into ICS.

Should founders buy treasuries/bonds/CDs with any of their cash? 

I’d generally say no for cash-burning tech companies. Keeping cash liquid tends to be important and there’s not a lot of incremental value for locking up cash and having to manage it vs. having an automated solution like insured cash sweep. That recommendation might change if you have multiple years of runway and meaningful cash ($100M+) and if the yield curve shifts and there’s a more meaningful return for locking up some portion of cash.

How much debt should I take on and when? 

This varies on the business type and stage. For early and early growth companies, you typically are doing a third of the last equity round and you’re doing it within six months of raising the round.  

What are good standard terms? 

This varies on the type of facility you are getting. For venture debt, good standard terms tend to be a third of the last round in size, four years overall term, no financial covenants or maybe a topline covenant, rate of Prime +0.50%; warrants in the 20bps fully diluted ownership range.

What are the gotchas? 

Know your lender and what kind of partner they are. If they don’t have a ton of experience, it might be risky to work with them as you don’t know how they will behave when it matters most.  Otherwise, I don’t think about it as gotchas. I think you just need to make sure the structure works well for the business. 

If there are covenants, do they provide founders sufficient flexibility and are they the right ones to manage the business with. How should founders pay it down?  

Use it when you need it. Pay it down per the terms of the debt deal unless you’ve raised a new round, in which case you should consider refinancing what you had with your lender.

When should founders consider working capital facilities vs. standard venture debt? 

This varies depending on the type of business and business model. Generally speaking, when revenues start getting to the $10 million range, you have enough working capital that can support working capital debt. Before then, you typically will have more availability through venture debt. Venture debt doesn’t scale well like working capital debt.

What about equipment financing/leasing? When should founders consider this? 

Typically when you have hard assets to finance AND it’s in excess of what a venture bank might be able to provide. Bank loans tend to be more attractive in how they are structured (cost and repayment).

I’m glad to connect with folks directly if they have questions or are looking for a better banking partner. You can reach me through email: mle@stifelbank.com.

---

Nate Bek is an associate at Ascend, where he screens new deal opportunities, conducts due diligence, and publishes research. Prior to that he was a startups and venture capital reporter at GeekWire.

Disclaimer: The information provided here is for educational and informational purposes only. It does not constitute financial advice, and you should always consult with a qualified financial professional before making any investment decisions. Past performance is not indicative of future results.

Token Talk: Open source won the AI race

By: Thomas Stahura

If it wasn’t clear already, open source won the AI race. 

 To recap: Deepseek R1 is an open-source reasoning model that was quietly launched during the 14 hours TikTok was banned. The reasoning version of Deepseek V3, Deepseek R1 performs at o1 levels on most benchmarks. Very impressive and was reportedly trained for just $6 million, though many are skeptical on those numbers. 

 By Monday, a week after R1 launched, the model caused a massive market selloff. Nvidia lost $500 billion in value (-17%), the biggest one-day selloff in US history, as the market adjusts to our new open-source reality.

 So, what does this mean? 

For starters, models have been commoditized. Well-performing open-source models at every scale are available. But that’s besides the point. Deepseek is trained on synthetic data generated by ChatGPT. Essentially extracting the weights of a closed model and open sourcing them. This eliminates the moats of OpenAI, Anthropic, and the other closed source AI labs.

 What perplexes me is why Nvidia got hit the hardest. The takes I’ve heard seem to suggest it’s the lower costs it took to train Deepseek that spooked the market. The thinking goes: LLMs become cheaper to train, so hyperscalers need fewer GPUs.

 The bulls, on the other hand, cite Jevons’ paradox. Wherein, the cheaper a valuable commodity becomes, the more it gets used.

 I seem to be somewhere in the middle. Lower costs are great for developers! But I have yet to see a useful token-heavy application. Well maybe web agents… I’ll cover those in another edition!

 I suspect the simple fact the model came out of China is what caused it to blow up. After all, there seems to be such moral panic over the implications on US AI sovereignty. And for good reasons.

 Over the weekend, I attended a hackathon hosted by Menlo where I built a browser agent. I had different LLMs take the pew research center political topology quiz. 

 Anthropic’s claude-sonnet-3.5, gpt-4o, o1, and llama got outsider left. Deepseek R1 and V3 got establishment liberals. Notably, R1 answered, “It would be acceptable if another country became as militarily powerful as the U.S.” 

During my testing, I found that Deepseek’s models would refuse to answer questions about Taiwan or Tiananmen square. In all fairness, most American models won’t answer questions about Palestine. Still, as these models are open and widely used and used by developers, there is fear that these biases will leak into AI products and services.

I’d like to think that this problem is solvable with fine-tuning. I suppose developers are playing with Deepseek’s weights as we speak! We’ll just have to find out in the next few weeks…

Investing in Oumi

By: Nate Bek & Kirby Winfield

Today, we’re excited to announce our investment in Oumi, an open-source AI platform. Venrock led the $10 million Seed round, joined by Plug and Play and Obvious Ventures. 

AI is at a critical juncture. Industry leaders are investing heavily in proprietary systems, with OpenAI’s $500 billion Stargate initiative serving as a clear example of the growing shift toward closed ecosystems. These walled gardens centralize innovation and resources, restricting access for much of the AI community.

Even within so-called open-source initiatives, transparency remains incomplete. Models like DeepSeek and Llama provide open weights, but fail to include critical elements such as full training pipelines, preprocessing code, and detailed data provenance. This lack of openness limits developers’ ability to reproduce results, refine models, or push the boundaries of what is possible. Without full access, innovation remains stifled, and the barriers to entry for AI development stay high.

A step-function change is needed to redefine what open-source AI can and should be. Oumi is addressing this challenge by building a fully open-source platform for large foundation models that spans the entire AI development lifecycle: data preprocessing, pretraining, fine-tuning, evaluation, and deployment. By providing access to every component, Oumi empowers researchers and developers to collaborate and innovate without restrictions.

Oumi partnered with more than 13 universities, engaged 36 AI researchers, and built a team of 10 high-profile builders to bring its vision to life.

Our conviction in Oumi was solidified through conversations with CEO Manos Koukoumidis. He has a track record of leadership at hyperscalers like Google, Meta, and Microsoft, paired with academic rigor as an MIT and Princeton PhD. Open collaboration on this scale requires not just technical excellence but strong leadership. Oussama Elachqar and the other seven co-founders were also stellar. It’s rare to find such an exceptional founding team on day one. 

“To protect their interests, closed vendors openly brag about how many GPUs they have in their attempt to secure their compute ‘moat,’” Manos says. “But I believe an open community can do at least as well as them. I actually think it can do a lot better.”

This vision aligns with and helps inform our AI Fast Follower thesis. The economics of AI development have shifted significantly. GPU arbitrage, open frameworks, and pre-trained models have slashed barriers for new entrants. Fast followers can achieve frontier-level model performance without relying on the deep pockets of the early movers.

That said, open source isn’t without its challenges. Monetization can be tricky, and aligning a decentralized community takes effort. But the potential payoff is enormous: a truly global effort driving better, safer, and more reliable models across every modality and use case.

Oumi’s platform is already addressing critical needs in industries where trust, reliability, and scalability are non-negotiable. This progress makes a strong case that open-source collaboration can deliver better, faster, and more inclusive innovation than walled gardens.

We’re excited to share the cap table with notable angel investors including Ruslan Salakhutdinov (Carnegie Mellon professor), Ragavan Srinivasan (Meta Product VP), Vipul Ved Prakash (Together.ai CEO), and Clem Delangue (Hugging Face CEO). 

Open-source AI has the potential to lift all boats, driving global progress and opportunity. Oumi embodies this belief, and we are proud to support their journey.

Meet Nazuk Thakkar, the Ripple Investor Who Lifts up Young Founders

By: Nate Bek

For young founders, Nazuk (Naz) Thakkar might be your strongest ally — literally. 

In college, she took over a powerlifting club and made it more open to women, sparking her passion for lifting and a drive to make traditionally exclusive spaces more inclusive. That experience shaped her mission: make hard-to-crack communities more accessible. It also gave her the grit to advocate for founders taking their first swing.

“It’s founder empathy,” she tells Ascend. “Running the powerlifting club and also being an active member in this traditionally male-dominated sport taught me how to support people building from the ground up.”

Naz also sees massive potential in young founders. “These young founders are fearless and committed,” she says. “They’re willing to take a bet on themselves, and that energy is what’s going to take their companies far.” 

As an investor at Ripple Ventures, a Canadian venture firm, Naz focuses on high-conviction pre-seed investments, writing checks between $250,000 and $1 million for core deals and $50,000 - $100,000 through Ripple’s Fellow Fund.

Ripple is currently investing out of its Fund III, anchored by the Royal Bank of Canada and the Government of Canada's renewed Venture Capital Catalyst Initiative ("VCCI") through BDC Capital. The firm focuses on companies tackling outdated industries like healthcare, logistics, and construction. 

Naz was kind enough to sit down with Ascend for our VC profile series, where we showcase early-stage investors from across the US. We talked in more depth about her  global roots, her views on young builders, and why sauce is the best ingredient. Read to the end for carve-outs.

*We've edited this conversation for brevity. Enjoy! — Nate 👾

---

Nate: Thanks so much for doing this, Naz. How did you get started on your journey of becoming a professional investor? 

Naz: I grew up in Ottawa, Canada’s capital, but it always felt like a small town. I literally grew up in farmland, surrounded by cows and horses. Back then, I didn’t know much about startups. My dad worked in cybersecurity, and when I was in fourth or fifth grade, he started his own company. Watching him scale the business and take that chance on himself was exciting and left a lasting impression on me.

Later, my family moved to Singapore, where I did high school. Living there broadened my perspective—I traveled to nearly 40 countries, which shaped my worldview. After high school, I returned to Canada to study business at Queen’s University. The standard career paths—investment banking, accounting, consulting—didn’t appeal to me, which was unsettling because everyone else seemed so sure of their career trajectory.

Some friends in private equity suggested looking into venture capital. It clicked. My dad’s tech background and my exposure to Singapore’s fast-paced innovation ecosystem made VC feel like a natural fit. At 18, I cold-emailed around 100 VC funds during my first semester. One fund, Trend Forward Capital in New York, took a chance on me. I spent 4–5 months there as an analyst, and it blew my mind. Venture capital was the most exciting and dynamic space I’d ever seen. Where I could be myself and thrive. 

You get back from New York and your internship, what do you do from there? 

Back at Queen’s, I helped launch the Ontario chapter of Front Row Ventures (Canada’s version of Dorm Room Fund) and created Queen’s first VC course to make this career path more accessible for students.

To gain operating experience, I became the first hire at a DTC brand in New York called Ruby. I worked on scaling their retail and customer experience operations, handling biz dev and sales. But I missed software and investing. That’s when I came across Ripple Ventures through their fellowship program. I loved their grassroots, operator-focused approach to investing. I pitched myself hard, and they brought me on. Four years later, I’m still here.

You mentioned your operator experience. I worked in retail too — at a fish spot and an ice cream shop. While that doesn’t exactly translate to software, I think there are some shared lessons, like dealing with black swan events and the unpredictability of running a business. As an operator-focused VC, what do you offer founders? Why do they want you on their cap table, both personally and as a fund?

Travel shaped how I work with people. Seeing different cultures and perspectives taught me how humans tick. At university, I accidentally minored in religion because I took so many courses to understand worldviews. That foundation helps build founder empathy and trust, which is critical at the pre-seed stage.

Founders face big risks and uncertainty, and I believe in being a good, kind human first. That’s why they want Ripple on their cap table. We show up—whether it’s answering late-night emails, reviewing decks at midnight, or just listening when things get tough.

It’s about being like a third co-founder, filling gaps and offering support. My own experiences—traveling, being an early operator, and facing challenges—drive that approach. I know how much it matters to have someone in your corner early on.

What are you personally excited about these days? App layer/infra etc.

I spend most of my time around application layer, digging into verticalized applications for cybersecurity and compliance. 

I care most about very clear, immediate and costly problems that are ripe for disruption and have budgets to access with unique distribution advantages. I’ve been seeing a lot of these trends in use cases within gov, pharma, healthcare, climate and fintech especially for example where we've done a few deals this year. 

You’ve done a few deals in Seattle, like Zealot, the spin-out from the University of Washington, and Climba, the agent company. Being based in Toronto, how do you position yourself in that market? And how do you approach competing in larger U.S. markets and winning allocation there?

Starting with Toronto, it’s about building rapport with founders. There’s a risk aversion here that’s more obvious than in SF or New York. But Toronto is getting to a place where founders are more comfortable. There’s a wealth of resources and a strong sense of community, similar to what I’ve seen in Seattle.

You see people co-investing and collaborating on deals. I’m passionate about building that ecosystem in Toronto. I host monthly VC events and spend time in the community building out our co-investor network. The grassroots approach matters. You need founders to trust you, even if they’re not raising today. You want to be their first call when they are.

Toronto founders often take the time to work closely with investors. But the brain drain is real. You have to win on value-add. That’s why Ripple focuses on pre-seed gaps — both capital and domain expertise. We invest in thought leadership, community work, and founder support. That’s where you win in Toronto.

It’s similar to Seattle. We travel to build our co-investor network and think about strategies to help portfolio companies scale. It’s about setting them up to win — not just in Canada, but globally. That means working with VCs who can back future rounds, building corp dev networks, and thinking strategically about what’s next for the companies to scale.

When I was in Seattle, I saw founders building in a space with gaps similar to Toronto’s pre-seed scene. Ascend and PSL are bridging those gaps, keeping founders from draining to SF. It’s all about post-investment work: graduating companies, co-investor networks, corp dev connections, and helping companies succeed beyond just your capital.

Let’s talk about Seattle. We’re a Seattle VC, and, yes, we pander to our audience — people click for that. You’ve been here, hosted an event, and made a few investments. Coming from across North America, what excites you about the opportunities here? AI talent stands out — that’s obvious. The legacy of juggernauts like Microsoft and Amazon is huge. But beyond that, let’s talk about the personality of the founders.

Our most recent investment in Seattle came from students and dropouts from UW. Making that investment, we realized there are incredibly smart kids here who are just building.

They have this “screw it, I’m just gonna build something” culture. It’s not as common in other universities and we need more of that self-commitment.

In Seattle, it’s different. These young founders are ruthless and fearless. They’re willing to take a bet on themselves, and that energy is what’s going to take their companies far. We saw that firsthand with our investment.

Interesting, I don’t hear that often. I’m friends with some of the most cracked young engineers here, and you’re right: they’re killers. Some that come to mind are Caleb, Jamari, and Parsa at our portfolio company Moondream. You alluded to it, but what’s the bear case? 

I think the ecosystem in Seattle, to talk about the bear case, needs to cater more to the young builders. There’s a very strong focus on investing in seasoned founders. We talked a bit about the access gap to capital for early-stage companies, but it absolutely exists for young builders too.

When I was trying to understand what organizations exist within UW and other universities in the PNW, there’s not a lot. But there are these kids who are so inspired. They’re close to SF, so they’re drinking from the firehose, surrounded by all the smart people in Seattle, and they’re willing to build companies.

We’ve spoken to so many who aren’t necessarily looking for capital but for support and value-add investors. The good thing is there’s amazing talent, hustle, and dedication to building here. That’s rare to find outside the traditional tech hubs.

People need to keep an eye on these young builders and pay attention to who’s up and coming, rather than always focusing on three-time exit founders or ex-Amazon execs. The folks coming out of school and building companies are some of the most committed I’ve seen. Turnover and commitment are always challenges for young builders, but what’s coming out of the Seattle schools is very impressive.

I don’t know how much you’ve been following what Kirby has been posting online or some of the things I’ve been involved with, but yeah, there’s a lot happening. There’s DubHacks, CSeed, and YoungTech Seattle. Switching gears, what songs are getting the most run on your headphones? 

My partner had never seen Twilight, so I was like, "Okay, it’s officially that time of the year — we have to watch all the movies." And we did. We binged every single one in a weekend. Since then, the soundtracks have been on repeat. I’m obsessed with "Eyes on Fire" by Blue Foundation —it’s such a classic. I also love all the EDM remixes of it. But in the gym, I’m in my punk rock era. And I’m a dubstep girl through and through — dubstep is my go-to for all my lifts.

What’s your go-to shoe? 

One, lifting shoes—absolutely. My heeled lifting shoes have changed my life. Seriously, I can’t imagine squatting without them. They give you depth, comfort, mobility and the range of motion you need. They’re 100% my ride-or-dies. Day to day, though, I love Nike Blazers. The color schemes are awesome, they look good with everything, and they’re just such a timeless casual shoe. 

Go-to ingredient in the kitchen? 

Oh, okay, I love this question. I’m a big believer that food is just a vessel for sauce. Like, literally, the whole side of my fridge is packed with different sauces—especially hot sauce. I really believe in the power of spice. I’ll add spice to anything. As for ingredients, the trifecta is garlic, onion, and chili. 

Honestly, you can’t go wrong with those.

Token Talk: Decentralizing AI Compute for Scalable Intelligence

By: Thomas Stahura

Compute is king in the age of AI. At least, that's what big tech wants you to believe. The truth is a little more complicated.

When you boil it down, AI inference is simply a very large set of multiplications. All computers do this kind of math all the time, so why can't any computer run a LLM or diffusion model?

It's all about scale. Model scale is the number of parameters (tunable neurons) in a model. Thanks to platforms like Hugging Face, developers now have access to very well performing open source models at every scale. From the small models like moondream2 (1.93b), and llama 3.2 (3b), to medium range ones like phi-4 (14b), and then the largest models like bloom (176b). These models can run on anything from a Raspberry pi to an A100 GPU server.

Sure, the smaller models take a performance hit, but only by 10-20% on most benchmarks. I got llama 3.2 (1b) to flawlessly generate and run a snake game in python. So why, then, do most developers rely on big tech to generate their tokens? The short answer is speed in performance. 

Models at the largest scale (100b+ like gpt4o and the such) perform best and cost the most. That will probably be true for a long time but maybe not forever. In my opinion, it would be good if everyone could contribute their compute to collectively run models at the largest scale. 

I am by no means the first person to have this idea.

Folding@home, launched October 2000 as a first-of-its-kind distributed computing project, aimed at simulating protein folding. The project reached its peak in 2020 during the pandemic, achieving 2.43 exaflops of compute by April of that year. That made it the first exaflop computing system ever.

This also exists in the generative AI community. Petals, a project made by BigScience (the same team behind bloom 176b), enables developers to run and finetune their large model in a distributed fashion. (Check out the live network here.) Nous Research has its DisTrO system (distributed training over the internet). (Check its status here.) And there are plenty of others like hivemind and exo. 

While there are so many examples of distributed compute systems, none have taken off for the reason that it's too difficult to join the network.

I’ve done some experimenting, and I think a solution to this could be using the browser to join the network and running inference using webllm in pure javascript. I will write more about my findings, so stay tuned.

If you are interested in this topic, email me! Thomas @ ascend dot vc

Token Talk: The Rise in Test Time Compute and Its Hidden Costs

By: Thomas Stahura

Reasoning models are branded as the next evolution of large language models (LLMs). And for good reason.

These models, like OpenAI’s o3 and High-Flyer’s DeepSeek, rely on test-time compute. Essentially, they think before speaking by writing their train of thought before producing a final answer. (This type of LLM is called a “reasoning model.”)

Reasoning models are showing terrific benchmark improvements! AI researchers (and the public at large) demand better performing models, and there are five ways to do so: data, training, scale, architecture, and inference. At this point, almost all public internet data is exhausted, models are trained at every size and scale, and transformers have dominated most architectures since 2017. This leaves inference, which, for the time being, seems to be improving AI test scores. 

OpenAI’s o3 nails an 87% on GPQA-D and achieves 75.5% on the ARC Prize (at a $10,000 compute limit). However, the true costs remain (as of Jan 2025) a topic of much discussion and speculation. Discussion on OpenAIs Dev Forum suggests, per query, roughly $60 for o3-mini and $600 for o3. Seems fair; however, whatever the costs are at the moment, OpenAIs research will likely be revealed, fueling competition, eventually lowering costs for all.

One question still lingers: How exactly did OpenAI make o3?

There exists no dataset on the internet of questions, logically sound steps, and correct answers. (Ok, maybe Chegg, but they might be going out of business.) Anyways, much of the data is theorized to be synthetic.

Image credit

StaR (Self-Taught Reasoner) is the subject of a research paper that suggests a technique to turn a regular LLM into a reasoning model. The paper calls for using an LLM to generate a dataset of rationals, then use that dataset to fine-tune the same LLM to become a reasoning model. StaR relies on a simple loop to make the dataset: generate rationales to answer many questions; if the generated answers are wrong, try again to generate a rationale given the correct answer; fine-tune on all the rationales that ultimately yielded correct answers; and repeat.

It's now 2025 and the AI world moves FAST. Many in the research community believe the future are models that can think outside of language. This is cutting-edge research as of today.

I plan to cover more as these papers progress, so stay tuned!

Mapping Seattle's Enterprise AI Startups

By: Nate Bek

At the start of the year, we spotlighted Seattle’s understated position as one of AI’s most influential hubs.

Since sharing that market map, the scene has evolved dramatically. A wave of AI startup formation and funding has followed breakthroughs in OpenAI’s GPT models and open-source architectures. Researchers, hyperscaler veterans, and serial founders are stepping away from established roles to launch new ventures.

Funding totals, as in other regions, are skewed by major deals. Dave Clark’s Auger ($100M) and Xaira Therapeutics ($1B). But beneath these outliers, a vibrant ecosystem of early-stage startups is thriving, often backed by top-tier investors

Founders are tackling niche infrastructure challenges and building app-based businesses. The city also spawned AI Tinkerers, now a global network connecting thousands of top AI innovators across more than 28 cities.

Seattle remains an enterprise-focused market. This has sharpened our focus on this market map to highlight B2B AI startups, open-source projects, and products, including: AI Stack, Model Development, Business Operations, Vertical Office Apps, and deskless workforce solutions.

Seattle remains the world’s second-most concentrated market for AI and software talent, behind only the Bay Area. The city’s hyperscalers are investing billions in new data centers, advanced chip development, and hiring top-tier AI professionals. At the same time, OpenAI, Anthropic, Nvidia, Meta, and Google have established satellite engineering hubs in the region, further solidifying its position as a global AI powerhouse.

Despite this concentration of talent, Seattle trails other metros in funding totals as access to local capital lags. Our focus remains on deepening our commitment to the region — fostering connections with local talent and supporting the next wave of transformative companies.

---

Methodology: Our map highlights companies, products, and projects we view as foundational to the broader development of AI, with a focus on enterprise applications, underlying infrastructure, and models. It features more than 120 logos showcasing the wide-ranging potential of AI in real-world settings. Bounding boxes identify companies we’ve invested in, acknowledging any potential bias upfront. This is not a comprehensive list or a ranking — rather, it’s a snapshot of the region’s evolving AI ecosystem.

Our 2025 Predictions: AI, space policy, and hoverboards

By: Nate Bek

Happy holidays from the Ascend team!

We enter 2025 with big questions on AI’s societal impact, new model breakthroughs, and a wildcard twist:

AI’s Societal Impact: How will AI reshape life as we know it? Will universal AI tutors enter classrooms? Will debates over AI ethics hit the Supreme Court? Will robots finally be… intimate? 

AI Technology Advancements: What’s next for the tech? Will AGI still be a pipe dream, or will we see a breakthrough? Will multimodal models dominate, or will agents take center stage? Could AI-powered apps become more profitable than the underlying models?

Black Swan Event: Anything goes — wild, random, unexpected. Will the Cascadia Subduction Zone finally do the thing? Will Froot Loops get banned under RFK? Will Spokane declare independence and start its own state with Idaho?

Here are our takes:

---

Kirby Winfield, Founding General Partner

AI’s Societal Impact: There will be at least one drone swarm attack on US soil in 2025. 

AI Technology Advancements: Test-time training will replace pre-training as the standard method of predictive model optimization.

Black Swan Event: The Mariners will win the World Series.

Jen Haller, Partner and Chief of Staff

Societal Impact: We will use AI tools to be more intentional about forming IRL connections with others.

Technology Advancements: This is less of a prediction and more of a wish: My Oura Ring tracks stress, heart rate, and sleep. By 2025, an AI agent should take that data and help manage my schedule. Shift meetings when I’m overloaded, remind me to take breaks, and keep me on track for a balanced day.

Black Swan: I finally get my hoverboard. 

Nate Bek, Associate

AI’s Societal Impact: Fans will appreciate availability over authenticity. AI-generated songs, movies, books, you name it — tuned to an artist's voice and style — will be good enough to be enjoyed by even the most loyal fans.

Technology Advancements: Small models will win the enterprise.

Black Swan: RFK will ban Froot Loops.

Thomas Stahura, Software Engineer

AI Societal Impact: Image-to-video will be the go-to meme format of 2025.

AI Technology Advancements: Coders will consume at least half of all generated tokens in 2025.

Black Swan Event: The outer space treaty of 1967 will be renegotiated.

Meet Tyler Churchill: From CAA to finding Startup Talent in Seattle for Bonfire Ventures

By: Nate Bek

Tyler Churchill  knows how to spot talent. Before he got into venture capital, he was at CAA, one of Hollywood’s top agencies.

“I definitely see a lot of parallels between CAA and VC,” Tyler tells Ascend. “Finding the next great actor is not so different from identifying a strong founder.”

More than anything, Tyler believes in a mentality of being a trusted partner in building and creating, stepping into whatever role is needed to ensure success. At the heart of this is understanding that the founder is the driving force, while he and others provide the scaffolding (cast and crew)  to bring their ambitious vision to life.

Bonfire Ventures is a Los Angeles-based Seed-stage venture fund focused on B2B startups, investing from its $168 million third fund. The firm manages more than $1 billion in assets. 

Tyler was kind enough to sit down with Ascend for our VC profile series, where we showcase early-stage investors from across the US. We talked in more depth about his VC passion, Bonfire’s focus on B2B applications, and what connects Hollywood to startup investing. Read to the end for carve-outs.

*We've edited this conversation for brevity. Enjoy! — Nate 👾

Nate: We’ve talked before about your move up from LA. I know you're representing Bonfire in the PNW, including your recent Supio deal—congrats, BTW! Can you share why and how you became a professional investor and what brought you to where you are now?

Tyler: I went to USC for undergrad, thinking I wanted to get into the business side of entertainment. I’ve always loved film, so after graduating, I joined CAA, working in the motion picture talent / lit departments. It was a super interesting experience, and after a couple of years, I went to work for one of our production company clients. I was there for a little over a year and got to work on the fourth season of Eastbound & Down with HBO. 

But while I was working in entertainment, most of my friends were moving to the Bay Area and getting into tech. The more I caught up with them, the more I wanted to make the jump into software. So I moved up to the Bay and got an early sales role at a bootstrapped software company called TechValidate, learning sales on the job. My experience at CAA translated well to software sales since both are very consultative and service-oriented.

TechValidate eventually sold to SurveyMonkey… How did that change your trajectory? 

TechValidate was eventually acquired by SurveyMonkey for nearly $100 million, and I stayed there for a while. But I missed the startup environment and wanted to learn more about building and scaling a company, so I joined a venture-backed startup called EverString. 

This was a big shift from TechValidate’s bootstrapped approach — EverString had raised a lot of money from Lightspeed and Sequoia and later raised a Series B. I helped build out the mid-market sales motion, and EverString eventually got acquired by ZoomInfo.

During this time, I got exposed to the VC world, hanging out with people in both startups and venture capital. I became fascinated by the shift from solving specific problems at a startup to addressing challenges across multiple startups as an investor. It seemed like a great way to apply a broader, more strategic lens.

Two acquisitions later, how did you manage to break into VC? 

I saw business school as a chance to manufacture an inflection point, so I went to Chicago Booth for my MBA. During the summer between my two years, I interned at Bonfire Ventures and kept working with them throughout my second year. I helped portfolio companies refine their models for Series A, sourced interesting companies, and stayed close to the team, showing my commitment to the firm. Eventually, I got an offer from Bonfire before finishing my MBA, and after graduating, I moved back to LA. I’ve been with Bonfire ever since and it’s been an awesome ride.

You invest at the seed stage, so let’s start with Bonfire's focus—specifically the verticals. I know it's primarily B2B software, but what are you focusing on now? Is it still the AI application layer, or are you looking beyond that, maybe into more traditional data? And what are you most excited about in the coming months?

Sure, I’ll break it down into the key areas we're focused on, and I'll speak for myself here too. Bonfire is entirely focused on B2B—though not just SaaS. We invest in FinTech and other B2B models, but our sweet spot is seed-stage B2B. For us, that means some early commercial traction, typically around $300K to $500K in annualized revenue. We usually write $2M to $3M checks, leading a $3M to $5M seed round.

We focus mainly on the application layer. While we’ve done dev tools, infrastructure, and API-based investments, most of our expertise lies in understanding market-specific pain points and how applications can solve those challenges and improve workflows. We’ve had a lot of success in vertical software, which is still a big focus area for us. We think there’s a significant opportunity here, especially in applying AI to specific industries where domain knowledge and understanding the nuances of end users and stakeholders are key.

We’re particularly excited about applied AI and how it integrates with vertical software. It’s not just about adding AI but knowing where it adds real value, which requires a deep understanding of users and customers. That’s where we feel we can make a difference—helping founders apply AI in ways that matter, refine the messaging around it, and build out the GTM processes and motions that help them reach short term milestones but more importantly lay the foundation for a healthy, scalable and efficient business. In vertical software it’s especially gratifying because many of these customers are SMBs or mom-and-pop businesses, the backbone of the economy, and most to none have the means of building things internally like software businesses. Helping them leverage AI and better software feels like a meaningful mission.

So we’re diving deep into vertical applications and applied AI across industries right now. It’s exciting work, and it feels like a great opportunity to make a real impact.

Jerry and Supio are great examples of that focus. Legal is such a clear use case for applied AI—there’s a lot of repetitive, time-consuming work that AI can streamline, especially for clerks and junior legal staff. It’s about making their workflows more efficient so they can focus on higher-value tasks. 
He represents the perfect founder archetype in vertical software where you need to be extremely customer-centric — listening to them, talking to them constantly, and certainly loving them. But you also need to become an expert yourself and be able to challenge them and introduce new ideas. Customers are great at telling you their problems - they are not great at articulating solutions. 

It’s not just about giving them what they say they want; you have to challenge them, too. That’s been an interesting dynamic to learn, and Jerry really embodies it well with incredible empathy and and energy.

I saw that firsthand when moderating the panel with Jerry. It’s interesting—I hadn’t heard “challenge and love” used with customers before, but I like it, especially for a tough industry like legal. It makes sense. Switching gears, what drew you to Seattle after LA, the Bay Area, and Chicago, beyond the nature? How do you see the technical talent and opportunities here, especially at the application layer?

The catalyst for moving here was family — my wife is from the area, and we always felt a pull to come back at some point. It wasn't clear when it would happen, but after our twins, Hayden and Charlotte, were born three and a half years ago, we realized we needed more space and family support. My team at Bonfire was very understanding and flexible, even though I’m the only one up here while the rest are in SoCal. It made sense for the business too — having a presence in the Pacific Northwest.

The ecosystem here is really exciting, with deep technical talent that’s no longer just concentrated at Microsoft and Amazon. There are many promising B2B software companies of substantial scale, and the technical and commercial talent pool is incredibly deep and growing quickly. The AI talent here is particularly strong, creating a lot of momentum in that category.

It’s a great time to be a seed investor in this region. We focus on later seed rounds but like to get to know founders early. Being here allows us to have those initial conversations, help with whiteboarding, introductions, etc and build relationships over time—so when they’re ready to raise a seed round, we’re already on board and there is a trust-based personal relationship underlying the diligence process. 

Do you have concerns about Seattle’s startup scene? Enterprise software dominates, and there’s talk of over-investment in AI, not a lot of venture capital, and limited go-to-market talent. What’s your take on the risks here?

I don’t think the concern around over-investment in AI CapEx is specific to Seattle; that’s more about where revenue opportunities will emerge to justify those asset valuations. My main concern is that Seattle’s ecosystem is still developing—similar to LA. It takes time and effort to build a network like San Francisco, New York, or Boston. And leveling up the region just takes dedication, intention and faith in what your community is capable of, but the PNW is incredibly deep here. People up here believe this can be among the biggest hubs for innovation, especially when it comes to AI.

The focus should be on creating networks where technical founders can meet go-to-market co-founders, innovative ideas can and chaotic serendipity can be fostered and harnessed. There’s a lot of energy and action in the veins of Seattle — look at what Aviel and others are creating with Foundations to build connectivity. People are fired up about it. 

Fun question time… What’s your favorite shoe? 

Running a lot and loving the Reebok Floatzig Symmetros.

What song is getting the most play? 

Still Hot by Nic D and Connor Price to get me going for early AM workouts, and anything from Zach Bryan. 

Love it. Zach Bryan’s “Sun to Me” has quickly become one of my favorite songs of all time. What ingredient is your favorite?

We eat a lot of salmon in our house, so probs have to go with that. And we have three kids now so fruit pouches I guess. 

Seattle!