First, we're going to just ground you guys. You hopefully have heard of these startups, but if you haven't, Mahi, could you just explain a little bit about what Higgsfield does?
So, we're a GenAI image and video platform. Our core customers are creators big and small. So might be a solopreneur, might be a marketer, somebody inside of a brand or somebody inside of an agency.
Very cool. And you guys recently raised a big funding round as well.
Well, we raised around the first week of December.
That's recent to me. So in AI terms, that's like a year ago. But yeah, the stats for the company: we launched our platform in April of last year, so not quite a year old. We're at $330 million run rate. 15 million users around the world.
Very cool. And I've played with it. Very cool tool. Another tool that I've played with that joins me in meetings is ReadAI. Rob, could you give people a sense of the scale of Read today?
Yeah. So, ReadAI, we built a meeting notes product. We've got five million monthly active users on that. More than 50% of our business is actually outside of the US. And the way that I think about what we're doing though is going beyond meeting notes, which is how can we remove bottlenecks and specifically for information workers, how can we make them not the information bottleneck? And so that looks a lot like either agentic or human-in-the-loop actions and things we can do with unstructured data after meetings and also incorporating email, Slacks and other sources of enterprise context.
Fantastic. Are you taking notes on this panel?
I'm not. Okay. It's a fair question.
Next time. Christina, why don't you tell folks about Linear if they're not familiar with what you guys do in product?
Yeah. So, we're a product development system for teams and agents. We have 25,000 plus paying customers, ranging from Ramp to OpenAI, Vercel, and many more.
Very cool. And this topic is really near and dear to my heart because Upstarts is a one-year-old ultra lean startup itself with some contractors and help and then I'm the only full-time employee after being in a large media org for 12 years. And I feel like I'm living firsthand what is the balance of growing fast but also efficiently as leanly as possible. And I'm selfishly going to learn from you guys because all your companies are much bigger than mine. But I am curious just to start maybe with you, Mahi. Higgsfield is only a couple years old but how have you guys kind of thought about what is the right balance of hyper-growth versus sustainable lean growth as well?
They're not mutually exclusive. First and foremost, I think the DNA of the company, we're an interesting mix of geeks, marketers, filmmakers, and customer success people. And I think in each of those categories we've been very careful about investing around our team around the same time we have product market fit feedback to give us scale. So we're 80 people around the world, we were 40 people six months ago. So, you know, we are scaling, but the amazing thing about AI is that you really don't need the same army of people, whether it's in product development or in go-to-market.
To be able to service this demand that's out there. At Read and Linear, which have both been around a bit longer, is there a pre-AI, post-AI cliff in terms of how you guys think about lean growth or what would the map look like for you, Christina?
Yeah. We're 130 people at this point. When I joined, you know, three years ago, we were maybe 35 people or so. And that was before this crazy AI wave. And in general, I would say we really aimed to be a lean company before it was popularized or before you could, you know, quote unquote outsource a lot of this work to agents or AI. And I think that's just a core belief from our founders that you should be a very focused team in the early days and then once you figure out what's working it's your opportunity to scale. So it's looking like this year we might be up to like 190 or something like that by the end of the year is probably around what we're targeting. So that's very fast growth for us in terms of what we've done historically. But perhaps it's the opposite approach. You're hiring all of these people when you do have all of this AI and technology to help empower them. But I also see it as there's a ton of opportunity in front of us right now. And I want us to be the company that takes advantage of that opportunity. And for some of that, sales in particular is not something that I think can be outsourced to AI fully at least when you're talking about going up market, spending more and more time with enterprise customers. That's just something that is fundamentally a people human business. So I think there are a lot of elements of our business right now where there are a lot of humans involved and going to continue to be. I see you nodding, Rob.
Yeah, I was gonna jump in there. I like to think about what's new and what's not new. And I think the thing that's not new is that every business has a bottleneck of some sort. Could be one function, discipline area, whatever it is, every business has a bottleneck. That's not new. What's new is a set of tools that can shift around where that bottleneck is. And so, is the bottleneck writing a PRD? Probably not. If certain tools are being used in certain ways, you can have a conversation. This is a shameless plug. You can have a conversation on ReadAI and then you can just ask it to turn it into a PRD. Great. Or your engineer can go into Cursor and say implement the thing we just talked about. Great. So then where's the bottleneck? And I totally agree with Christina that the bottleneck is going to be and should be if the tools are used correctly, the areas that necessarily need a human. And I think sales is a great example of that, which is that all of this automation and acceleration that happens in the day-to-day tactical information worker doesn't apply to human touch necessarily. It can help you do a better job at human touch by retaining context and familiarity and awareness, but it's not going to replace that one-to-one human time. And so when I think about how we've grown, it just changes, I think, over time where we view the bottleneck and where the sense of urgency is to hire. We're about 90 people now and we've been doubling every six to eight months for the recent quarters. So that's how I frame it. I think about bottlenecks. Bottlenecks are something that's always existed. And then I think the bottleneck is also the missing piece of the conversation if I can bring in pop culture news and things with Block laying off 40%. What I'd love to hear is, okay, so what's the bottleneck now? And what was the bottleneck to actually using that 40% of the workforce? Why can't you just grow faster? And I think that's sort of the unsaid piece, which is what would it have taken to not have to do that and have growth be 10x what it is today? Or is the bottleneck just your TAM at this point? And having those explicit conversations is useful.
Well, the Block example could be a bit of an extreme one. You know, it obviously Block tripled its headcount in the previous four or five years before they laid off 40% of people. Maybe in the more middle ground, we've seen companies like Shopify say our approach is going to be prove to me that AI and our current software or something we vibe code can't do this thing and then you can hire the new person, but you kind of have to justify each new hire. Do you guys think that is a healthier middle ground or what would be sort of the approach internally for you?
I mean, when I joined Linear that was the initial approach and this was a kind of pre-AI world. And I think most of the companies that I think are very thoughtful with hiring should have that back and forth with the team or the functional leader that wants to hire versus saying hey you get this headcount envelope and marketing can go and hire 20 people and then the marketing leader is like okay I guess I'll have to go figure out what all those people are going to do. Like, I think you really should be advocating for every single role. And there should have always, even pre-AI, been a question of saying, okay, is there a reason this can't be part of someone else's job? Is there a reason tooling can't ultimately solve this problem and now it's AI tooling or just general tooling can solve this problem for us or is this actually a job that maybe should fit into everyone's scope versus being something specific? And I think those are the kinds of questions you should be asking to ensure that you're not just hiring to empire build or do things that a lot of executives do when they typically join companies and want to grow teams.
Interesting. I mean, some of the people here are investors. This is an investor-heavy conference. If one of their companies was saying, hey, we want to grow cautiously, focus on profitability, that might not be exciting to investors who want a power law outcome, right? Like huge growth. And then we also see startups talking about, hey, hire for the inflection point, like the inflection curve, hire ahead of the need and then maybe that person grows into the role. So, how do you kind of bring in the right person who wants to maybe grow into a role who you don't necessarily know exactly what potential they have and also at the same time be thoughtful because obviously no one intentionally overhires, but then we do see startups sometimes end up with that 20 person marketing team or that five open-claw engineers that they're not really sure what to do with.
Rob, you want to maybe start? Yeah, I think what's old is new again. I think the intention is never to overhire. I think that it's a process failure on the part of companies to not have the reflection on why and to keep plans that are stale for too long. And so if you've got a hiring plan that's looking out 12 months and you're not revisiting it, maybe that's a side effect of some of the tooling and some of the velocity changes that we've seen. That plan really is only high confidence for one quarter and you need to talk about what the bottlenecks are going to be after that quarter. And to the investor piece, what needs to happen for capital to be the bottleneck is a question, right? It's not the case that everything can be done without capital just because I can buy a Mac Mini and set up a bot. But we just have to have all those conversations and figure out what does the landscape look like because the problems still exist. They've shifted. We've pushed on the balloon. Something else is popping out and we have to reflect on that and I think finding the people and putting them at the right spots in the organization who can reflect on that is the hard part at the end of the day. And it's easy for me sitting here to say well Block should talk about what their next bottleneck is and figure out how to use those people. It's incredibly hard though and I completely concede it at an organization that size to have that context. So, at our company sizes, I know sort of by gut what's going on and where there are bottlenecks and where there are not. And I'm sure you do too, Christina. But, at many thousands of people, it's a different story. And if you don't have the visibility, then you're flying blind and you're at risk of doing that.
Well, Mahi, I'm curious. You're in this very competitive AI video space where top AI talent come at a power law premium themselves. And working with models can get really expensive. Have there been intentional places where Higgsfield was like this is where we really want to commit resources. This is what is going to be a 10x priority compared to others where maybe we can be lean and mean.
Yeah, it's a great question. One of the things I think part of this is driven by our founder DNA. We are split between San Francisco and Almaty, Kazakhstan. So we have access to world-class ML engineers and researchers that we're paying a fraction of what that would cost in San Francisco.
How did you guys pick that location?
Our CEO was born in Uzbekistan and our CTO is Kazakh. So homefield advantage. Absolutely. And the thing that a lot of companies that start in San Francisco don't necessarily have that same DNA, but they should look at other geographies because there is really strong talent outside of the Bay Area. The Bay Area is unique and that's why we're HQ'd there, but it's also a very tough place to recruit top tier talent. I think the other thing I'd add to that is having a, we have not just an international workforce, we have a 24/7 working clock and that gives us just immense scalability and the fact that that international workforce allowed us to build products that are more easily accessible from all over the world that we don't sell to a North America kind of taste.
Do you balance intentionally where you do the hiring or do you set out with a global plan for the company?
I think we've been more opportunistic but as we've grown our team in Central Asia it's become a recruiting hub for us.
Yeah. You know, it's interesting. On the next episode of our podcast coming out tomorrow, actually, I spoke to the CEO of Synthesia, which for those who don't know is another AI unicorn. They do enterprise video, so not really what you guys do quite, but also in the video world broadly. They have, I think, offices now in like Ljubljana in Central Europe, and their hub is London, and then they of course have people in San Francisco and New York, too. But they've increasingly been hiring engineers in the Balkans in Central Europe and it feels like that has become maybe a playbook again. It reminds me of some of SaaS businesses hiring talent in some of these places 10, 15 years ago.
Yeah, if I think about for Linear maybe we think about this mostly as an engineering thing, right? And we hire engineers all over the world, but primarily in North America and Europe just for time zone and syncing purposes. But outside of engineering, half of our marketing team is in Europe, right? And this is talent that would be very difficult to find if we were solely looking in San Francisco or even in the United States.
Are they cheaper on a per headcount basis or is it just more efficient to have marketers in Europe?
I mean we would hire that marketer wherever they are. So it is not an intentional like we are hiring for this specific role in Europe. We basically say we will hire for this role anywhere in North America or Europe and great candidates find us or we find them and they happen to be in a different place and yes the cost is cheaper compared to I would say tier one cities specifically in the US but in general I would say it's probably less about the cost of talent and we're a profitable company, have been for years, so it's less of our concern. It's more that a lot of this talent is overlooked by other companies. Companies in the Bay Area want to hire marketers and salespeople and a lot of go-to-market folks in the Bay Area or at least in the United States and don't necessarily look at that as a talent hub to recruit from. So, a lot of these candidates are amazing, but they don't necessarily have access to the same number of opportunities. And you can deliver them a great opportunity at a great company and they're super excited to work here.
I'm curious, are you guys all in office cultures? Like...
That was going to be my question for the other companies because Read is very...
I'm sorry. There's a good question though. You should continue.
Answer first. Are you guys in office at Read?
We are. So, Read, we're based out of Seattle. And probably 75% of the company is in Seattle. And I think we have almost the opposite effect which is there is a lot of hunger for people who will move to Seattle for Amazon or Microsoft or whatever the case may be. They want to get into a smaller company environment and they want it to be something that's found product market fit and they can learn and in many cases build their own companies one day. And so we can really take advantage of that feeder network flowing into the region and find some great talent there where there's less competition. So we do in-office most days a week. In fact, the people that we talk to typically are relieved. They like that. An engineer three to five years into his or her career hasn't had an in-office job potentially. And standing at a whiteboard with somebody is a thrill when you get to do it on a regular basis and you've been looking for it. So we are pretty in-office focused.
