Drill Down Ep. #256: QumulusAI (QMLS:NASDAQ) Building the Bottleneck Michael Maniscalco

Cory Johnson (0:09)

Welcome to The Drill Down, we've got the business stories behind Stocks In A Move. I'm Cory Johnson, Chief Market Strategist for Epistrophe Capital Research. Well, six weeks after going public, this AI infrastructure company has signed a big multi-million dollar deal. Now the CEO explains where the capacity for AI is going to come from. But first, it's sponsor time. The Drill Down is brought to you by Aera. Never miss another critical insight or event ever with Aera. Customize your company watch lists, track key events, mentions, filings, and more, all with an easy to use customizable interface. That's Aera, aiera.com.

I'm Cory Johnson, welcome to The Drill Down, we explain the business stories behind Stocks In A Move. It's first quarter of the public company, there's AI infrastructure firm. Fascinating, QumulusAI's CEO joins us right now, Michael Maniscalco.

Michael, thanks for, you go by Mike or Michael?

Michael Maniscalco (1:00)

Mike's great.

Cory Johnson (1:00)

I see Michael in the documents, I see Mike as sort of...

Michael Maniscalco (1:03)

He's good, yeah.

Cory Johnson (1:03)

There we go. Well, Mike, glad to have you here.

I put QumulusAI into the world of NeoClouds, is that accurate?

Michael Maniscalco (1:12)

Yeah, a new school of clouds is how I would define what we're providing today, with a lens on AI first rather than general compute. So yeah, I would say that when I think about QumulusAI and where we fit and how we describe it, NeoClouds are a perfect representation.

Cory Johnson (1:26)

So talk to me about the business. How do you sort of go after this business? We certainly read about the big projects, and I spent a lot of time working on understanding the data center projects at AWS, or a lot of time working on Oracle and so on, you know, Hut 8 in the news as we record this, but talk to me about what you guys do and where your niche is within the world of AI.

Michael Maniscalco (1:48)

Yeah, so generally, when you're looking at the Neo Clouds, like I said, it's AI first cloud.

So looking at compute, storage, networking, and power required to bring that type of workload infrastructure online. So what we do is we provide compute access for our customers, and our customers are looking for that compute as quickly as possible to keep up with their growing customer demand. So our customers typically look at like the AI type companies or enterprises adopting AI at scale, and they're seeing a lot of success as these AI use cases take off, and that success and their ability to manage that success is driven by their ability to acquire infrastructure and compute. And doing that is oftentimes challenging because it requires a lot of capital, a lot of power, a lot of power to sell our data center space, as well as expertise to deploy and operate. So what QMLS does is we package that up and turn it over to our users' turnkey infrastructure they can run their AI applications on top of.

Cory Johnson (2:39)

And why would a customer pick you instead of, I don't know, going to... We've heard stories about Meta looking at renting out their capacity, other bigger Neo Clouds out there, you know, Nebius and the like, or even going to an XAI or Grok, which has been a lot of headlines renting out space that they weren't able to use with Grok.

Michael Maniscalco (3:02)

Yeah, so this is a challenging space to plan. So when our customers are coming to us, they're generally coming to us because they can't find the compute they need and the timelines they needed or the pricing they needed.

And that's just the reality of the market we're in today. So when you look at a lot of the hyperscalers and you look at the hyperscalers earning calls and even our peers earning calls and you're hearing their comments that we are growth constrained by capacity. That's the theme you hear across the industry. So what QMLS has done is found different ways to bring capacity online that better aligns with the pace that AI is moving. So think months, not years of developing massive data center capacity to serve.

Cory Johnson (3:37)

Yes, faster. Yeah, everyone wants it now, right?

Michael Maniscalco (3:40)

Speed is a key tenet of ours, right? Speed with some levels of durability. Velocity is a big piece of our decision-making process because that's what our customers want.

Cory Johnson (3:48)

Yeah, it's interesting, because again, I spent a lot of time with this stuff. I was at a dinner last week with a whole bunch, I guess it was two weeks ago, with a whole bunch of venture capitalists, probably a hundred different VCs. And I was kind of working the room in the pre-gaming of the dinner, just trying to get a sense of whether their companies, the companies these investors have invested in, are still so hungry for access to compute. And they are, at least, you know, that was the takeaway I had from this event. Again, I probably talked to 50 VCs, and they told me every single one of them concurred that their companies they've invested in are desperately trying to find more compute. That's a good environment for you.