No Priors Ep. 142 | With Harvey Co-Founder and President Gabe Pereyra

1. Key Themes

The Evolution from Individual Productivity to Organizational Transformation

Harvey's journey reflects a fundamental shift in AI application from individual lawyer tools to firm-wide business transformation. Gabe explains: "In the past year and going forward, the big problem we're solving is not, how do you make individual lawyers more productive? It's how do you make a team of lawyers working on a client matter more productive? And more importantly, how do you make an entire law firm working on thousands of these client matters more productive and more profitable?" [00:01:38] This represents a strategic pivot that has enabled Harvey to capture significantly more value and embed itself more deeply into customer operations.

Legal Work as Complex Reasoning Traces, Not Just Document Processing

Legal work parallels software engineering in its complexity and decision-making processes. Gabe draws this comparison explicitly: "When that translates to an RL environment, part of what is missing from the public models is the process of looking at one of these entities and figuring out given all of the context of I want to do this merger. This is the right way to structure it. Just that process and a lot of the reason in trace." [00:16:02] The critical insight is that public models only see final SEC filings from mergers, missing the entire reasoning process that led to those decisions - similar to seeing final code without understanding the architectural decisions behind it.

The Professional Services Platform Vision Beyond Legal

Harvey is positioning itself not just as legal software but as infrastructure for all professional services collaboration. Gabe articulates this ambitious scope: "For us, the bigger opportunity seems how do we build the platform that lets professional service providers and their clients collaborate? And I think a lot of the problems you need to solve there are... The biggest is the secure collaboration across many of these entities, the secure data sharing." [00:28:41] The addressable market expands from a $1 trillion legal market to a $3-5 trillion professional services market.

2. Contrarian Perspectives

AI Will Make Organizations Grow, Not Shrink

While conventional wisdom suggests AI will lead to downsizing, Gabe believes the opposite: "I think the interesting thing will be the transition from like these models are really smart individually. But if you think about like a lot of what we've done in the past 20 years with SaaS, it's how do we use software to make these massive organizations? And I think that will be the continued trend where a lot of what we're starting to think about is like law firms have like 10x in size compared to before computers and the internet. And I think that's going to happen again." [00:41:31] This challenges the narrative that AI primarily replaces workers, suggesting instead it enables organizational scaling.

Building a Law Firm Would Be the Wrong Strategy

Despite frequent suggestions to compete with customers by building an AI-native law firm, Gabe firmly rejects this: "I think the big issue you run into if you try to do both of these is I think you can only do one thing well and doing a law firm well is very different than building a software company well... I think the real problem we're trying to solve is can we make every law firm more profitable?" [00:26:32] This contrarian view prioritizes platform leverage over vertical integration, betting that enabling all firms creates more value than building one proprietary firm.

The Partner Role Won't Change Much, Actually

Counter to expectations that senior roles will be disrupted, Gabe argues: "I don't think the role of law firm partners actually doesn't change that much. In the same way, I don't think the role of very senior engineers changes with this because you're largely delegating work. And what you're getting paid to do is here's the high level strategy. Here's the right abstractions." [00:13:14] This suggests AI transforms execution layers while preserving the value of strategic experience and client relationships.

3. Companies Identified

Cursor

Description: AI-powered integrated development environment for programming Why mentioned: Example of finding the right form factor for AI coding tools that emerged later but achieved product-market fit Quote: "I think with coding, the initial models were also not quite as good that you needed maybe a bit more capabilities of the base models. And then you needed, I think figuring out the right way to like integrate this into the, into the idea." [00:36:49]

Sierra

Description: Company with agent engineering program for customer service Why mentioned: Comparison point for Harvey's field deployment engineering approach Quote: "I would say this is closer to like Sierra's agent engineering program, but what we're starting to run into a lot is I think early on we did a really good job of building a horizontal platform." [00:20:17]

A&O (Allen & Overy)

Description: Major international law firm Why mentioned: Harvey's first customer that rapidly scaled from pilot to firm-wide deployment Quote: "Our first customer we actually met through you introduced to an ex partner who was doing business school here and he introduced to David Wakeling at A&O. I was in our first year and they went from a small pilot to firm wide and investing in this." [00:25:00]

PWC (PricewaterhouseCoopers)

Description: Global professional services firm Why mentioned: Example of large enterprise customer requiring significant customization Quote: "Then for very large customers like PWC, we did some customization." [00:20:59]

Walmart

Description: Major retail corporation Why mentioned: Recent major enterprise customer win for in-house legal department Quote: "So we recently announced we signed Walmart. We're working with AT&T. A bunch of these Fortune 500 large private equity firms, Global 2000, kind of the largest consumers of legal services." [00:02:37]

Glowball (Gloval)

Description: Fast-growing private equity firm Why mentioned: Example of PE firm working closely with Harvey to develop new workflows Quote: "Glouall is one of the fastest growing private equity firms that we recently started working with. We meet with them all the time and there's all these things that we feel like we could map into Genai." [00:22:00]

4. People Identified

Gordon Moody

Description: Former partner at Wachtell Lipton, now advisor to Harvey Why mentioned: Exemplifies the highly technical nature of senior legal expertise, comparable to distinguished engineers Quote: "Gordon Moody was a partner at walktale, which is one of the top transactional firms in the world that joined us really on and is now an advisor... he was a part of when Michael Dell took Dell private and then restructured it and took it public again. And this was like a multi year super complex financial and legal restructuring of an incredibly large business." [00:14:22]

Winston (co-founder)

Description: Harvey's co-founder, former first-year associate with exceptional business intuition Why mentioned: His unique combination of legal expertise and strategic thinking was crucial to Harvey's founding Quote: "Even though he was a first year associate, just had this intuition of not just the work he was doing but the structure of the firm... He was in the process of convincing some partners to leave to start a law firm with him, which is insane for like." [00:32:58]

David Wakeling

Description: Contact at A&O who became Harvey's first customer Why mentioned: Instrumental in Harvey's first enterprise deployment Quote: "Our first customer we actually met through you introduced to an ex partner who was doing business school here and he introduced to David Wakeling at A&O." [00:25:00]

5. Operating Insights

The Field Deployment Engineering Model for Enterprise AI

Harvey created a "deployed engineering force" to embed with customers and drive adoption, learning from enterprise software playbooks. Gabe explains: "There is a massive amount of demand of we just want smart technical people to sit here and help us think about our business and our operations and how we should start mapping that into Genai systems. For us, it's a really good way to figure out the roadmap." [00:21:47] This approach accelerates product development by turning implementation into research and development, creating a flywheel between customer needs and product capabilities.

Building for the Future Capability, Not Current Limitations

Harvey designed their product architecture assuming rapid model improvement rather than optimizing for current model constraints. This forward-looking approach enabled them to avoid building narrow solutions that would quickly become obsolete. Gabe notes: "I think that was something we did really well where we just had this belief where I think the same that you see with the programming products where if you had built something where it's like, all this does is like check that your Python code doesn't have bugs, which you could have done better with 3.5. Like you wouldn't have built something like cursor." [00:34:10]

Document Upload Plus Citations as the Killer Feature

The initial breakthrough feature was deceptively simple but addressed core legal workflows. Gabe recalls: "The initial feature we built that none of the products had at the time was upload a document and do something with it, right? And that is a lot of legal tasks. And it was that and then do really accurate citations. And when you showed people that, they were like, oh, this is crazy because that's so much of my job." [00:36:32] This demonstrates the importance of identifying the atomic unit of value in a new domain rather than building comprehensive solutions immediately.

Law Firms as Implementation Partners

Harvey discovered that law firms themselves want to become implementation partners for their clients. Gabe describes this emerging ecosystem: "Law firms are starting to do this for their in-house clients. They're starting to go and take Harvey and go to their clients and say, hey, buy Harvey and we'll help you build all the workflows and implement it because we have the scale and the expertise to build this." [00:23:17] This creates a powerful distribution model where customers sell to other customers.

6. Overlooked Insights

The Partner Feedback Loop as Proprietary Training Data

The most valuable data for legal AI isn't public cases or documents—it's the internal feedback partners give associates on their work. Gabe briefly mentions: "If you think of what that reward function is at the law firms, it's the partners, right? Okay, there is no way to verify this besides the senior partner who's done a bunch of these said, yeah, this looks pretty good. And so internally, these law firms have a bunch of data of here's all the edits that went into this and the feedback." [00:18:28] This subtle point reveals that Harvey's true competitive moat may be access to proprietary reasoning traces and expert feedback that will never be available to public models, creating an increasingly defensible data advantage.

The Verification Problem Will Persist Even in "Verifiable" Domains

While discussing why legal is hard to verify compared to coding, Gabe makes a profound observation that even coding verification is illusory at scale: "I think you actually have the same problem in programming where I think in the short term programming is verifiable where you can look at unit tests. But once you get into real software engineering, like the unit, there is no unit test. It's like I deployed it. It's like I deployed this and a million users used it for six months and it didn't crash." [00:19:01] This suggests that the distinction between "verifiable" domains like code/math and "unverifiable" domains like legal/writing may be less significant than commonly believed, and that all complex professional work ultimately relies on long-term human judgment rather than immediate programmatic verification. This has major implications for where RL can be successfully applied.