Ep 67: Max Junestrand (CEO, Legora) on Differentiating and Pricing AI Apps & How the Legal Industry Will Evolve

Today's unsupervised learning was a particularly fun one. We had on one of my portfolio company CEOs, Max Unistron, who is the CEO and co-founder of Legora. Gore is at the forefront of applying AI to the law industry. They're working with many of the top law firms around the world. They've raised over $100 million and honestly one of the fastest growing AI applications out there. I was joined by my colleague, Logan Bartlett, and the three of us hit on all sorts of things, including...

Max's take on the future of... the law space, how he thinks about product prioritization, given the rapidly improving models, and what he thinks about building versus leaving to the models. And we also talked about what it's like expanding across a bunch of different global markets. This was an awesome conversation. Without further ado, here's Max.

Max, thanks so much for coming on the pod. Yeah, thanks for having me. We've been excited to do this one for a while. So have I. It's great to be here from Stockholm. Yeah, we've got all our Red Point podcast hosts today.

Yeah, exactly. That's my first appearance on Unsupervised Learning. Yeah, we've tried to keep you out for a long time, but you're destined to wear a statue. You don't want to water down the quality that is this podcast. Yeah, exactly. I appreciate you having me on. We'll try and do like an NIL crossover episode.

Yeah, you're keeping me in my little box on my podcast. Well, Max, a ton of stuff I know our listeners will be eager to hear about. Maybe to start, can you just contextualize for us? Where are we in the AI meets law space today? What works? What doesn't? Right. Well, when we started, nothing was working. We were using the earlier... What was that?

The early BERT models from Google, and they were decent in English, horrendously bad in Swedish. And this was back in 2020. So when GPT arrived 3.5, that was like the paradigm starter, if you will.

Since then, I think we've moved from full-on experimentation and trying to get stuff to work into actually implementing things that are really taking end-to-end work deliverables. Just to give one example, if you're doing a due diligence today, you're not going physically into the data room. You're not using Control-F. You're just taking all the documents, putting them in Legora, saying what you want to find, and then it finds it. And then based on those findings, we generate the report.

And so things are really starting to move from empty queries against a data set to, OK, this is the process that we want the LLM to follow. And what we do is we give an agent access to tools. It then plans, executes.

on that plan using the tools, and then we get an end-to-end work deliverable that's actually usable. And where do you think this is all going? Obviously, these models get better every three, six months. If we were to ask you to have a crystal ball and say where the legal field's going in five, ten years, what do you see ahead?

Also, the interesting thing with the models getting better, that's one piece. But what we're getting the most leverage out of is actually all these surrounding frameworks like function calling, tool calling, MCP.

um the legal software space has been incredibly fragmented that was like one of the early things i saw coming from outside the industry you had one tool for translations one tool for document comparisons another tool for searching and another one for

viewing and now suddenly all of this is kind of getting baked together. And you can of course also imagine a scale of sort of the complexity of legal work and at the bottom you have something very simple like just... finding data extraction and at the top you have something really complex like drafting a share purchase agreement and we've already started to fully automate a lot of the bottom

quartile in this graph, right? And we're sort of slowly but surely moving up. And I think here, the really interesting thing is thing is for for law firms and legal professionals to see where are they going to add the most leverage and where do we need their specific expertise their context their way of instructing the models and where are the sort of

off-the-shelf LLMs just plainly good enough as they are. I'm sure we have a lot of people listening that are aware of the legal field and the implications for it, but can you maybe speak to some of the elements? of law and why it's uniquely well suited for ai as a potential application yeah yeah sure so um i i don't come from the law space i was an engineer before and

When we first arrived, I think the first thing that you notice is there hasn't really been that much software developed in this space. There's industry-specific incentives that... perhaps doesn't always align and incentivizes being more efficient and using software. So basically the hottest thing or the coolest thing that you could do was use a templating system. And what you basically have is this dilemma of law firms and in-house councils and they do very different things in-house councils

primarily work with the same stuff over and over again. It might be NDA reviews, MSAs. You're controlling the risk for the business. And when you work with a law firm, they typically do more one-offs or... complex projects or things where there might not be as much precedent and so on. And I think very broadly, you can sort of stitch it up in reviewing, reading, drafting, writing, or researching.

In the beginning, I think every piece of software in this industry was very focused on one of these. But because AI is capable of doing things across this entire stack, platforms like Legora have emerged that are not... solving a point solution but rather serving this entire wall to wall of of needs and and what we've really found is law firms want to lead in this new you know

paradigm, this new future. And for law firms who are not really leaning in, I think you kind of risk two things. One is not upskilling your team in this new future and this new paradigm in terms of how you're supposed to work. But increasingly so, clients are putting pressure.

on law firms to make this shift because they're starting to use tools like this internally a lot of ceos you see on twitter and linkedin posting about we're gonna go ai first right and and you can't motivate um give you can't motivate getting a new head count unless you can prove to us that you've been more efficient. I think it was actually the Fiverr CEO who also mentioned Legora for their legal stack. And I think that is...

To me like the most interesting thing when you start to blur the edge of what is software and what is service Because really interestingly in the legal business Software is like a 20 billion dollar market and legal services is like a trillion dollar market

Yeah, it's interesting. I feel like so many people thought this like hourly billing problem would be would make it really hard to adopt software. But I think you're totally right that, you know, if it's something that your clients are using, you need to be on top of it as well.

And maybe clarify that point. The hourly billing problem is a sort of an incentive-based... Yeah, that basically, you know, if the pricing models I bill for every hour I do, if you make me 50% more efficient, I'm billing for 50% less hours. But it's like...

I feel like lawyers used to manually look up things in libraries, and now they have databases. There's definitely been examples of things that have been adopted. For sure. And one thing that maybe it's not as clear from outside of the industry is that there's a ton of write-offs.

and price pressure. A due diligence used to be really expensive and now it's almost starting to get to the point where clients aren't even willing to pay for it um if you're a large pe client of a law firm um you will for sure pay for the advisory to the board when you're buying a company but you might not pay for the contract review and already you see

you know the large american firms outsourcing this work because it it isn't profitable for them to staff an associate who's billing 800 bucks an hour on a task like that and it's a bit of a prisoner's dilemma in that if any of your

competitors do something, you have to adjust to keep up with the pace of play. So even if you're theoretically have the mindset that you're talking about, if someone else does it, then you're certainly suddenly inefficient or you're billing for things that other people want. That's exactly it. You have an almost perfect equilibrium of the service, and as soon as somebody moves down...

it forces everybody to do it because there's very low differentiation on particular tasks like that. I think we want to dive into a little bit of the AI within Logora, but I guess one question just at a business level is there's other... players, unfortunately, in the space that you're operating in. I'm curious what advantages you feel like you've had both coming from the Nordics and also...

maybe starting six months 12 months after some of the other players in the space what does that advantage you as a as a company so i think starting in the nordics it's almost like we we were this really really small fish and we got to

eat our way to become a bit of a bigger fish. And then we suddenly became a crocodile in this smaller pond. And now we're jumping across the Atlantic, we're coming to the States, we just opened in New York. And frankly, the great thing was Europe is such a fragmented market in every country, effectively.

our initial market was very small. That was one of the first things that we got pushed on during Y Combinator. It was like, how are you going to move from Sweden? And I said, well, we're not. We're going to do Sweden, we're going to win it, and then we'll move. And funny enough, I think one of the main advantages being the sort of fast second mover, if you will, is you can observe what's working, what's not working. I think initially there was a lot of focus on we need to train our own LLMs.

We need to be an AI company first that's pushing the research in this field and coming second with also significantly less funding. I mean, our initial angel round was $50,000. And we just said, hey. We don't have the money to do that. We don't have the time. We don't have the energy. Let's just build an application and let's focus on the application layer and build something that at the end of the day, people are really excited to use. And coming from a non-legal background...

it also forces you to be very humble and be very attentive to what everybody's saying. So what are the clients saying? How is the relationship between a law firm and their clients developing? And I think this is also...

now pushing us ahead where we can move from being just an internal facing tool into really focusing on the entire relationship that exists in the delivery of high quality legal work. I did a podcast one time with Daniel Leck and he said starting in the Nordics allowed him Had he started in the U.S., they would have made him...

uh skinny down his product a little bit into a more narrow functionality but because he started in this other pond he could be a little bit more ambitious unknowingly because he serviced the totality of the needs of the market i'm curious of that if that's something you felt

I don't think I've thought about it that way, but I think if you look back on it, it kind of explains what we did, right? There's a lot of legal tech companies that are solving a very niche problem. Drafting. Drafting, contract review, or they solve search.

It might be super, super narrow. And even if you look at a big law firm, you've got litigation, you've got corporate, you've got the transactional team, and they work very differently. And we sort of said, hey, we want to service every lawyer. I want every lawyer who's serious about doing great work and making money to use Legora, just as every great designer uses Figma. And perhaps you wouldn't be that ambitious had you started here because of the competitive pressure, right?

i think there's also one nice thing about being able to serve enterprise in your local market to the point where you're already enterprise ready when you enter a new market and law firms are very well connected if you're a large firm in sweden you have a almost partner firm or a or a good friend firm in in france and in germany and in spain and in the us so by starting to work with a couple of firms you also get really great referrals because they all

you know, want to lean in and kind of do it the same way. I mean, I think one thing that's really powerful about your product, as you just alluded to, is just, you know, how broad the capabilities are. And you really are able to serve kind of end-to-end at these law firms. I know a lot of builders out there are kind of...

grappling with some problems, you know, or thinking through what happens when I have this end-to-end product? How do I like teach people to use it or how to get started with it? And I'm wondering like what you've learned, you know, now with all these deployments you've done about like how to teach lawyers to use these tools.

Well, I think the short learning is it takes way more work than we thought it would. One great example is Jem, who comes from Baker McKenzie. He was responsible for the Gen.AI rollout. And it was funny talking to him because in the initial... you know traditional software rollouts he did you would be thrilled to get five to ten percent adoption like that would be a great number but for any other enterprise software that is awful like those metrics suck

And now with Legora, when he's doing other client deployments for us, we're increasingly hitting numbers like 70% to 80% adoption. And it's a different world when the lawyers are... actively approaching the innovation team saying, hey, we want access. We want these tools. And that hasn't really happened in this industry before. How do you think about...

build towards with the models constantly improving. You want to skate to where the puck is going in some ways, but you also have to be pragmatic about what's actually applicable today. And so how do you think about that balance? And at what point you sort of think about product roadmap meeting model?

You can think about the models, but I also think it's useful to think about these AI labs as platforms and software companies. Because frankly, OpenAI, Claude, Gemini, they're for sure model providers. They're also increasingly product companies. You know, Anthropic is building out Cloud to connect with other systems. Google is building out Gemini to sort of sit on top of the entire Google workspace stack.

they're not only pushing innovation on the models themselves, but also on the way that they interact with other pieces of the software and system. So the way that I've always thought about it is if it's something that the AI labs are going to build and... at some point make available to builders like us, then we should not build it.

totally against the stream and you want to build everything like boats so that when the tide rises you know all the functionality just gets better but a great example would be We just released a new feature called Playbooks, where you kind of give Legora a set of rules and typical fallbacks and points of how it's supposed to negotiate. And let's say you outline 20 different rules for how to negotiate an NDA or a MSA.

Now, LeGuard needs to go through them one by one by one to make really high quality edits. But if the models become so good that you just say, hey, here's a...

playbook in a Word document or in an Excel file, take this other NDA and cross-reference them and give me all the red lines, then you don't really need to build the feature that way. So that's something where we think it... it adds a lot of value building it that way today but five years from now maybe that's completely unnecessary yeah it's such an interesting tension where it feels like

you want to provide value to your customers today with whatever the scaffolding you need around the models, but also realizing that some of that scaffolding may just be completely irrelevant in two, three years. Exactly. And I think another great example would be workflows, like multi-step instructions.

Typically, you would have these sort of node-based, no-code builders where you have the output of one block serving as the input for the second block. And you'd need these kind of technical builders to come in and set up these workflows. the LLMs themselves with proper planning and then execution with tool use are fully capable of on the fly creating their own workflow.

or like their own plan of it and so the only thing you need to do is really provide it with an instruction and say hey i want you to do this task and then it's going to be able to figure it out themselves given the tools that are at its disposal I realize there's no perfect way of predicting what the model capabilities are going to be even tomorrow, let alone 6, 12.

months from now if you can let us know yeah that would make our job a lot easier if you figure that out but uh i i guess at a practical level that that point that you made about like what exactly can be done potentially in the future versus how you

you need to do it today. Do you have some framework that like, hey, if it really feels like this could be done by an LLM in the next three months or six months, we might, let's hold off, let's table it versus it feels way in the future. Let's just go build it. Like, how do you think about that? You just saw... Cloud and Enthropic release citations as part of their API now.

We built citations really, really early on because it's critical for our use case and for lawyers to be able to reference and see, okay, the LLM made this response and it references to this material and this specific text chunk. If the LLM providers give that for free in half a year, then we'll just deprecate our entire code. My way of thinking about it is, if somebody else is doing it, then let's use that. What's the hardest part of building these products today?

i think the hardest part is um kind of to what logan was alluding to there's there's a hundred different things that you kind of want to be doing and they're all really high value and then you need to prioritize like which five are we going to go after and then how do these five

tie into each other in a very cohesive platform, because it's easy to build a Frankenstein monster if you just always go after like the next hot thing. And I think some other companies in our space has made that mistake of just

building stuff that kind of makes sense but the totality and the sum of it isn't really like well thought through so it's hard to plan for how your platform is going to look like and be structured just from even like a data data um you know a database perspective where okay we have an organization and then we have projects and these projects can maybe be shared with a client like how is how is all of this going to look like and how do we build the functionality so that that fits

the paradigm for what we're going for. And here as well, I think the law firms themselves and the in-house teams are kind of figuring out what they want as well. So we're all trying to figure out exactly the best way to apply the technology, and then we need to align on a future that makes everybody win. Some people put the cost structure in some ways on the cloud.

that you map, hey, this is what it costs us and therefore this is how we're going to bill you. That's not how you guys bill today. I'm curious how you think about... the the pricing and the value and the cost and how those three things kind of all interact and therefore what that means as model costs

continue to decrease? How do you pass that on to the customer? Yeah, this is a great question. I was talking with Varun from Clay about it in a cab the other day, and we have the problem of running a seat-based model, which I think is easy to buy. It's easy to predict. You know what you get. But then just the other week, we had a user rack up $10,000 in LLM costs. So I think over time, it's not hard for me to imagine where you'd have kind of a

platform fee and some usage element to that. But it also depends on how the LLM providers themselves develop their pricing. I think the early thoughts around all of this was LLM prices will continue to go down. And so if LLM prices continue to go down, we're willing to take a bit of a hit in the beginning to get positive and better margins over time. Now that's not really happening because the LLM models are becoming so much better.

but also more expensive. I don't know if you've tried 03, but 03 is incredibly good at a lot of legal work. it's incredibly incredibly good but it's also very very expensive and so i do think that there's kind of an element to this where it's like do you need the the bazooka where you just need a handgun for the task that we're doing. And increasingly so, we're running classification algorithms and model pickers that kind of pick the best model for the task.

I think it's such a fascinating point because it's probably next year, that same kind of recreation might actually be quite cheap, but it'll also be so last year. And there'll be some other model capability that's even more powerful that folks will end up wanting. Exactly.

What about on the infrastructure side? You're obviously one of the biggest users of LLMs out there today. How do you think about some of the gaps in infrastructure tooling that exists, some of the stuff you've used that you found helpful? The most interesting thing to me is MCP, where we can give Legora access to call on outside tools.

We can provide a set of tools that are already good, and let's say we give Legora access to Redline and Documents, and then that's a tool that the LLM can use. But the really interesting part becomes when our clients can provide tools as well. So you'll have Legora access a client-specific CRM or a client-specific knowledge database or a set of templates or...

you know, it's able to push notifications to the client's emails or something like this, right? And that sort of very, very quickly expands the universe for what's possible. Since we're a venture-sponsored podcast, I have to ask you every VC's favorite question, which is, how do you think about what the moats are in AI applications? The moats, of course, look very differently across industries.

Take Figma. I think the moat in Figma is that it's kind of the system of record and it's the platform where both the designers, PMs, marketers are collaborating, but also where they're... directly collaborating with their clients. And I think for us, there's a lot of interesting directions to explore there. And we currently integrate it on top of a lot of our customers.

own data we integrate with outside databases for things like case law legislation up and coming regulation and a lot of it actually sits in the system that you'd like to work in, I think. It's sort of a taste-based, I like working with this thing. I think if you grab the average lawyer in big law, they spend 80% of their time in Microsoft Word. outlook and i manage their document management system and i think that's about to change

Yeah, no, it's funny. This is one we love to debate internally. And I feel like it's always the question of like, is this just literally the same stuff as SaaS? And I feel like in many ways, in the early days of AI apps,

As you mentioned earlier, people were training their own models, they were doing all these things that felt like, oh, that's going to be super differentiating. And then quietly, I feel like all that's gone away and people have realized not as relevant. No, for sure. And I mean, look, there were... a lot of companies in our space too that said that there's the the moat is within fine-tuning models and we were very clear from day one and said there's no mode at all there and we're gonna

work with every model to build the best system. How do you think about building the system and the architecture that gives you the ability to decide this model here, this model there, and make sure it's flexible and future-proof? I feel like I'm just doing advertisement for a lot of the tools that we use internally, but one of them is Braintrust, where we've set up thousands of evals, and it's extremely easy to run an evaluation on a new model as soon as it becomes available.

And very importantly for our clients is that every model we use goes through security, privacy, legal review, so that it conforms with our data processing agreements. When you work with law firms, you've got to be on the right side of your contract.

The way that we've structured a lot of it is, let's say you have multi-step workflows, we then have evals for each step, and then we have evals for the complete end-to-end product, and we test a lot of the permutations around using the different models in different places. And those really high-quality reasoning models are incredibly good. They're just very hard to price.

When a new model comes out, do you have like a go-to thing you test? Or put it another way, is there like some capability you can't wait? Like some prompt, you're like, I'm so excited for the day that this thing works. Wow, that's a good question. The sort of most complex thing that... you might look to solve in a one-off prompt would be drafting a really long and complicated document. And the reason why that's hard is, you know, a comma there.

or a word there can directly influence the meaning quite a lot. And many of these contracts are not open source or not available on the web. So the LLM... don't necessarily get trained on them, right? Like, yeah, you have stuff on SEC and EDGAR, but a lot of the, you know, share purchase agreements that, you know, end up sort of being the driving precedent, they're firm specific. And so...

The models aren't necessarily getting trained on how to draft really high complex contracts like that. But then again, we can give a model or an agent access to a tool to access a repository to use that as the basis for a draft that it does. I guess shifting gears to, you know, uh, Lagora the company. Um, I'm curious, like, obviously you've had, you know, kind of a fascinating journey. And so maybe to hit some different parts of it, you know, you obviously were in YC in this, like.

post-ChatGPT craze. Like, what was that like? Was it kind of, I mean, now I'd say it feels like every ambitious founder is building an AI company. Did that feel in the air at the time? Or what was it like being in some of those, you know, some of those batches right as this craze was starting? So we...

We were, I think, one of the first acceptance into our batch. So we got accepted in August for the winter batch. So it was like an early... We got an early, and as soon as we got in... I think this is very unconventional and maybe not something I recommend, but I took out a loan against our investment coming in from YC because we had a Swedish corp and had to flip it to a Delaware and that was going to take like three months and we needed the money now.

just to hire four more engineers to run at this. So I think the feeling very much was the world is moving really fast. and we need to do so as well and when we got to yc of course everybody was working with ai it felt like or they were either building like a vector database or like a postgres you know layer on top of it and or they were building an application layer. I didn't take part in YC that much.

because I was doing all of our commercials at the time, so I bought this ring light to put on my laptop, like a thing that influencers have. And I was up between 1am and 10am for five, six days straight just doing sales to Europe. Then I kind of crashed and I said, fuck this, I'm going back to Stockholm. And the ring light, to be clear, was so it didn't look like it was the middle of the night. Exactly, yeah.

But then I would sit in the kitchen. I think we have some really fun photos of me and Tuhin, our first growth hire, sitting there and just crunching. And frankly, I think a lot of the YC companies... They build a lot. They're just very, very focused on building. We were very, very commercially focused from the beginning. And I think we really took the advice to heart that launch as soon as you have something. And to be frank, the first thing we had was...

and basically a private and compliant chat GPT with a better rag on their own documents and Swedish legislation. But that was good enough for that time. And then, you know, every single week, the bar increases.

for where you need to be to be best in class. I'm curious because this idea of launches as soon as you have something. I feel like in this AI wave, we're seeing people make really big promises ahead of where products are and selling maybe the furthest ahead that we've seen in quite some time.

And I think one thing you've been like very methodical about when it made sense to, you know, release product. Like I remember right after our investment, you were like, we want to do more work on this product before we bring it into other markets. Like how have you thought about that? Like tension of, you know, God, this market's moving so fast to.

Hey, I get one first chance. Five million. And, you know, look these guys in the eyes and say, yeah, we're not going to sell for the next, you know, four or five months. We're like, great. Yeah, I got some looks from Logan. It was a very mature decision, albeit as a first board meeting. It's the first time I've ever had that happen in my venture career. It's worked out. And I think to your question earlier, Jacob, you get one chance. You get one chance, especially with...

I don't know if this is especially with lawyers, but I say developers, if you're building vibe coding software, they're easier to approach once and then say, oh, we have some new updates. Do you want to come and try again? Because they like that idea of... You're starting somewhere and you're building and you get to come along the journey. Now for us, if an attorney comes in and they do a query and it doesn't work out, then we see them fall off. They don't come back.

And so, reactivating them is really hard. I mean, in the beginning, there were a lot of just reliability, infrastructure, just the way that you do chunking and, you know, you get the rag system to deliver at a certain expectation and then just being able to serve like thousands of users on the platform a day. Those were problems you had to solve in the beginning. And if you don't solve them and you try to onboard a firm like Cleary Gottlieb.

it's not going to go so well. And as we've moved from serving large firms in Europe into now serving some of the largest firms in the world, the expectations are incredibly high. we're also starting to see that we have become a system of record and a system that is more and more frequently being relied on for the end-to-end deliverable. If something isn't working, now we get...

you know, immediate phone calls, emails, Slack messages, like, hey, we need this thing up and running because we have a client deliverable to get to here. So I think, yeah, that was a mature choice. But now that we've...

Now that we're a bit bigger, we're not doing this big spring release and winter release yet because, frankly, everybody wants more all the time. But we've moved from... ship it as soon as we have something into let's work with a couple of design partners get it to the point where they're you know thrilled about it and then do a proper launch i'm curious like the The pace that you move at is seemingly the pace that AI moves at, which is...

I don't know, 100 miles an hour and just all over the place, things changing all the time. I call you, you're in one city, then the next day you're in the next city, and then the next day you're in the next city. And all those are three different continents that you're crossing or something.

But I'm curious how... you're able to get that to manifest within the business itself. You have this innate drive and hunger to do all that. How does that trickle down into the business? How do you make sure that that manifests itself in the go to market team, the product team, the engineering team and all those things?

I think in Stockholm, because we have engineering and product base there, and then we have three commercial hubs with New York, London, and Stockholm as well. You had Klarna, you had Spotify, you've had a couple of... larger companies come out of there and then it's been quite quiet for a long time like there hasn't been that much action you know web web 3 never really like took off and then you had ai and and i think you had

a lot of people in in the city that want to work really hard for a mission and a product and building a company that they believe in and so for us we've been able to attract i think you know the players with the most urgency to like really get shit done and that also comes from i think my co-founder sigge like i've never seen somebody just sit and code as much as he does it's like 14 hours a day he's in the office

coding and then it does one hour of climbing in the morning to stay sane you know seven days a week and i think if if you continue to run at that pace and the whole company sees that it trickles down and We've also been very upfront with that. So when we recruit, we're very upfront with this is not a nine-to-five job. We're not here to maintain something. We're here to build for the future.

if you're excited about that hop on board if not you know i'm sure there's another great company that you can go and work at I think you set that expectation with us pretty clearly. I remember, I think the first time we ever chatted, you were in Estonia at a conference. It was like 11 p.m. and you were drinking a Red Bull. Yeah. I feel like that's, if we were to introduce you in a novel, that's how I feel like your character would be introduced. Yeah, and it was fun.

When we signed the term sheet with Iconic, I think Seth was over at the office at like 1.30 a.m. And me and Sigurd were still there working. We had a conversation in December where it's like, hey, we have one or two paths. We can go down. We can just... fortify the flank and dominate europe or we can go to the u.s and you almost cut me off before i finished uh the question and said no there's no question we're going to go to the u.s i guess what have you found in shifting from a european

business and largely based in Stockholm, a little bit in London, which gosh, I mean, six months ago we were, which is amazing where we are today, to moving at hyperspeed, coming into the US and being able to balance. the culture and try to make all that stuff work together. So we force everybody to do like a week or two weeks of onboarding in Stockholm. And we've been very diligent on no remote work.

we're fully in office you know it's like momentum breeds momentum you know people love winning and i think we've been able to recruit people who love winning and who hate losing and so the losses feel really hard and and frankly For the past three or four months, we haven't had that many losses, which is nice. But in the beginning, I think every...

Every customer in the beginning that came back with negative feedback or when we lost a deal because we weren't seen as big enough or a big enough name, that really hurt. And now it's felt... great to go back and even having them reach out to us saying, hey, we just saw you raised your Series B, can we please have another conversation? I think we've very quickly moved from being a small Swedish-based startup. A year ago, we were 10 people. Now we're 100.

And recruiting 90 people in a year when you're that few and getting it right, that's hard. I think this part is so interesting. I mean, I feel like there was kind of a typical growth rate of a company that we all got used to in the SaaS era, and it's like, all right, you build for a while, then maybe you'd get a big corporate logo, and then maybe you get a few more. And it feels like you're, you know...

probably at the Vanguard, but there's a lot of AI companies that are having this kind of growth rate and employees and, you know, getting to these insane customers and, you know, pretty short order. I'm sure you talk to a lot of other founders, like, you know, maybe from the previous generation.

To what extent, like, does, you know, is company building feel different in like this kind of pace? And to what extent, you know, you know, we talk about sometimes like, oh, an AI series C company can feel like a series A company at its bones. Like, how is that, you know, maybe reflect on that.

yeah i think you're completely right in that um there's a new expectation for how quickly a software company can grow and i think a big part of that is you're not going in to replace an existing piece of software you're creating a completely new category. That is different. And you've had times like this in the past during the internet and during mobile where things were just possible that you couldn't build before. And so whoever builds the fastest...

has the highest velocity, the best product, the best service, they're going to end up dominating the market. And the really cool thing for us was our first client, Mannheimer Swartling, and our first design partner, they were an enterprise client.

They're a huge company. They're the biggest law firm in the Nordics, and they take really seriously on the work that they deliver. And so by sort of osmosis, I think that colors off, and you don't... you know you're not on this journey where you start serving other startups and then you start serving smbs and then you do you know you dip your toes in enterprise we were like day one enterprise and we spent

I don't know if I told you this, we spent half of our initial angel funding on SOC and ISO certifications. And I didn't take a salary for the first six months so that we can actually afford. doing that. Well, we always like to end our interviews with a standard set of quickfire questions where we squeeze in a bunch of overly broad questions right at the end. To start, what's one thing you think's overhyped and one thing you think's underhyped in the AI world today? I think...

MCP is both under-hyped and over-hyped. I think it's under-hyped in the sense that it's making it possible to serve a universal app.

into a lot of other different capabilities but it's also completely overhyped in the sense that everybody's you know talking about it and trying to just poc it and it's It's still not yet moved its way into full production. You kind of need to get some things right around authentication and other things.

What's been the biggest surprise like in building AI features at LaGora? Maybe one thing that you didn't think would work that actually worked quite well or one thing you were so confident would work and didn't quite take. the the most surprising thing is actually how people start using it um like when we dropped our drafting thing the first thing that happened was an attorney comes in and they just give a query which is write me an spa and i'm like

Of course that's not going to work. What are you expecting here? So I think the expectation management is surprising how it's, yeah. How do you teach people how to use it? Exactly, exactly. But that's why we do all these trainings. That's why we do the onboardings. That's why we do office meetings with all these big firms that we go around to. It has surprised me that

people come in with so different expectations. Because you can have associates who are super savvy and they know exactly how they're going to set up the workflow, they have their templates, their prompt library, and they're just so on it. And then you have folks coming in and they expect the world.

Yeah. That's hard. Well, I think one of these things that's interesting probably is, you know, as these model capabilities change every six months, you probably have to go back and be like, well, actually, now maybe you can write that query. Yes. Yes. I've thought a lot about how we can productize the onboarding of these things. What's something you changed your mind on?

uh since you've started lagora our first the first version of the product you've never seen this that was summer 2023 it was completely a button and a workflow based So you would have basically a library of nine different buttons in different colors. And one of them would be called summarize. The other one would be called check.

document against policy and another one would be called search a database so you had to go and click them and then you'd follow a very structured flow um one of my co-founders was very you know in love with that idea and i said no let's do the chat because i thought the chat would give us a better interface to do more and then we could always add these things as again like functions or tools or things that the chat can then use and so moving

From that to the other world, it basically meant that we deleted 95% of our source code as soon as we got accepted into YC, which I think was like a... We've taken a lot of contrarian moves. I think that was also one of the good pivots we made. If you weren't in building a legal space and you were kind of just getting started on something new now, like what other ideas in AI get you excited? I'm so in the legal world that it's like hard to think outside of it.

But just some of the things that I'm passionate about outside of this. I think CROs in pharma has like the biggest... It's one of the biggest disruption opportunities in the world because it's so manual. There's so much data. One of my family members works with clinical trials. They pay billions. billions to these CROs, which are effectively slow-moving consultancies that have, frankly, pretty structured workflows.

fully AI CRO that can ensure and deliver end-to-end deliverables, they will do really well.

I like that. Obviously, a lot of Legora is text-based today. Any multimodal use cases that get you excited? Video, audio, image? One of the things I'm most excited about is just starting to work with voice and audio transcripts, both in the sense that you can instruct Legora in voice, but also use things like like transcripts, audio files, because if you look at a typical deposition, like that's being recorded and...

Now you might previously have to hire somebody to sit down and note that down, which then serves as the input for wherever you take the work. But now just uploading the audio files, transcribing them and starting to work with them as... documents and interrogating them is, I think, a great use case.

Obviously, your tool can do a lot of legal work. As folks think about going into a career in law, what skills will matter for a lawyer in the future? If you're in law school right now, for those listening, what should they focus on? I've spoken with a lot of managing partners and management groups on this because they're thinking really hard about how they're going to upskill and train their new associates and the lawyer of the future.

As a side note, one of the funniest usage... peaks that we've had was when a Spanish firm had their whole new cohort come in and they all became like Legora fans that day one because it was part of the onboarding and then the usage for that firm just like spiked. But previously you might have hired underconfident overachievers that are really good at following instructions. They're very thorough and they sort of do things.

step by step by step. I think now you're going to need entrepreneurial, creative people who maybe challenge the existing ways things have been done because you will have partners who are, you know, sitting on top of their pyramid kind of farming and doing work there. a lot of the processes within six to 12 months, I think will be not disrupted, but augmented with AI elements. And I think you do want people who were upskilled or who were very fluent in working with AI during their...

legal studies. Yeah. I mean, it's really interesting. It's like, you know, everyone's going to be a manager from day one of a bunch of AI agents. And so it actually is a very different skill set than being a diligent associate. Completely. And you could say, hey, we think it's the innovations, innovation departments or the...

department's job to do that but I think it's frankly up to every individual to kind of augment themselves with this and I think that's starting to become very clear in the sense that you see all this again like Twitter and LinkedIn pieces coming up from tech CEOs Like, hey, you need to augment your own work with AI and show us how you're doing that. I think we'll have expectations like that in law firms as well.

Well, Max, this has been a fascinating conversation. I'm sure folks will want to go learn more about Legora, the work you're doing. The mic is yours. Where can folks go to learn more? So since we changed the name from Leia, we've actually bought the .com. So you can find us at legora.com.

I think the future is just too exciting to be left to just software engineers, and we really want to collaborate with law firms and legal professionals in building this future. So please reach out, and I would love to chat. Amazing. Well, Max, thanks so much for coming in. This is great. Thanks for having me, guys.