Who owns the model? Where should a legal team's intelligence live? These questions are at the center of many conversations in legal AI, but we think the most important question to answer is: what produces the best outcome for every legal task? As CTO @jacsebl and CPO Bryan Tsao explain, there is no best model. Different models lead on different tasks, and the frontier changes almost weekly. At Legora, we use the best available model for each task, and invest in the system, where intelligence compounds and remains editable, auditable, and portable. We post-train when we know it delivers our customers better performance on a specialized task. Training is a tool, not a strategy. No lawyer should have to worry about which model did the work. Just whether the work holds up.

I first came to San Francisco when I was 15, with my dad. He is a programmer. To a teenager from a small island outside of Stockholm, this was the place where the future was being built. We walked into Salesforce East and stood there watching the enormous digital display of moving water. Outside, Salesforce Tower was starting to rise. I remember looking up and thinking it would be pretty sweet to have a tower someday. We are not there yet. What I could not have known is that I would be back years later, having dinner with @Benioff and welcoming @salesforce as a @WeAreLegora customer. Salesforce's Legal and Corporate Affairs organization will now use Legora across North America, EMEA and APAC. Their lawyers keep the judgment calls. We take care of the work around them. We have run on Slack since day one and became Salesforce customers when our sales team outgrew its first CRM, so it is good to have it run in the other direction too. Welcome to Legora, Salesforce. Dad, this one feels special.

We are putting a new foundation under legal research in @WeareLegora. Comprehensive data, a full ontology of law, and an AI-native citator. Almost every legal question starts in the same place. What does the law say, and does it still hold. That is also where AI has been least worth trusting, and I think it is the hardest problem in legal AI. Two things have to be true: You have to have the law, and you have to know your way around it. Getting the data is a grind, and a different grind in every country. We partner with publishers where we can. Where nobody will partner, we go and get it ourselves. Manual requests, physical scanning, whatever that jurisdiction takes. We are working through over 100 countries and every type of source. Then the harder half. No agent can reason across hundreds of millions of documents. Something has to choose the sources before the reasoning starts. So we are building an ontology of the law and an AI native citator, compressing corpora of thousands to hundreds of millions of documents into structured data that an agent actually can use to provide reliable output. The publishers did this by hand. 150 years, thousands of attorney editors, every opinion read by a person. We have hired the best of those editors. They set the standard and they call the close ones, but AI does the muscle work. Reading 60 million pages of case law is no longer too expensive to attempt. The ontology is in limited beta now. Generally available in Q4. Full story:

Five numbers tell you whether an AI business is a real business. Gross retention.@WeAreLegora is 95%. Customers who bought last year are still here this year. If this one is broken, nothing downstream matters. NRR. Ours is 300%+. Gross retention is the floor. NRR is how much taller customers build on top of it. We don't sell shelf-ware.  DAU/MAU. Ours is north of 50%, and the average active user spends 17 hours a month in the product. A rollout tells you a firm has signed. This indicates the work actually matters, and it moves here before it appears in retention or NRR. Win-rates. Our August pilot closed-won-win rate was 78%. Winning roughly 4 out of 5 competitive pilots is downstream from offering a superior product. Gross margin. The one that matters most. Ours is positive and improving every quarter. Our customers want Legora to be a long-term partner, and this is what makes that possible. There's a shorter route: Price below what it costs to serve, book the logo, and hop on a never-ending fundraise treadmill to pay for it. The top line goes up, everyone claps, and every new customer costs more than they pay. Scaling a negative margin only exaggerates the problem. The whole point of scale is that the margin improves as you go up. Ours does. That's the only version of this business worth building.

Jake and the entire Legora engineering team has built remarkable technology from harnesses, routers, and more. Excited to see Jake share more publicly about the underlying technology that has powered Legora to be one of the fastest growing software companies of all time!

Performance and $$ depend hugely on harness design. In vertical AI, the gains from a great specialized harness >>> gains from more expensive model inb4 kimi has quirks - @runta can you try with a diff model also

The story of ClickHouse is truly insane. Started as an open-source project; scaled into the fastest-growing database product ever. Year 1: $0 Year 2: $12M Year 3: $50M Year 4: $200M Year 5 (not complete): My bet is $450M. My notes from our discussion with @ceo_clickhouse below 👇 1. Are Large U.S. Enterprises Scared to Work With Frontier Model Providers? Large enterprises remain skeptical of “zero data retention” claims and wary of sending proprietary source code to frontier labs due to IP indemnification and data leakage concerns. Rather than exposing production code, companies may limit frontier model usage to less sensitive workflows like code review while turning to open-weight alternatives for critical data. 2. How Do You Assess Defensibility and Moat in Companies That Scale Faster Than Ever Before? When an application scales from zero to $100M in ARR in a single year, investors must rigorously question its underlying moat. Hypergrowth without high switching costs leaves companies vulnerable to rapid churn as customers move effortlessly to the next model or tool that leapfrogs the incumbent. 3. How Does This AI Cycle Compare to Prior Technology Shifts and Transitions? Unlike the gradual adoption curves of the internet and mobile eras, the current AI wave is accelerating at an unprecedented pace. Agentic experiences are maturing rapidly, driving explosive revenue growth and placing historically unique performance demands on underlying data infrastructure. 4. What Job Does Not Exist Today That Will Be Very Prevalent in Five Years? A critical new corporate role could be an AI finance function dedicated entirely to managing token consumption and resource allocation across the enterprise. But the role may ultimately be short-lived as autonomous AI agents increasingly manage their own infrastructure spend and budget execution. 5. What Should Investors Be Worried About Today That They Are Not? The biggest overlooked risk in AI today is revenue durability. While infrastructure software benefits from high switching costs, agentic applications can have exceptionally low barriers to switching, raising questions about long-term retention as models and products continually leapfrog one another. 6. Why Revenue Concentration Is a Real Concern Operators and investors should treat revenue concentration as a critical risk, with any single customer or vertical accounting for more than 10% of revenue representing significant exposure. Sustainable enterprise value requires a diversified customer base so losing one account never threatens the company’s overall growth trajectory. (links in comments)

Contrarians tend to do very well at Ramp. @shevchenkoaalex, who runs @ramplabs, is one of them. High-agency people doing hard work should have real say in what work is worth chasing. Especially when everyone else says "that sounds too hard" or "that'll never work." Those are usually the projects worth trying. If you hear "impossible" and enjoy proving people wrong through action, you're exactly who we want building with us at @tryramp:

Why did a spend management platform start its own AI research lab? @karimatiyeh describes @RampLabs as a “collection of crazy ambitious people that don’t like to be told that something is impossible”. It started about a year ago and has since worked on projects like building its own production-grounded coding benchmark “Ramp SWE-Bench”, putting Claude Code in RollerCoaster Tycoon, and a mechanistic interpretability playground. The Redpoint team spent the day with Ramp Labs to answer questions like: ▪️ How did Ramp Labs get started? ▪️ How do they decide what research to pursue? ▪️ The long-term vision for Ramp Labs