{"slug": "how-sandstone-grew-40x-in-147-days-on-vercel", "title": "How Sandstone grew 40x in 147 days on Vercel", "summary": "Sandstone, a legal workflow automation platform, grew its revenue 40x in less than six months after launch on Vercel, handling over 1,000 legal requests daily across customer teams. The company built multi-step agentic legal workflows end-to-end using Vercel's AI SDK, with a 7-app turborepo monorepo deployed to multiple Vercel projects. Co-Founder and CTO Liam Germain credited Vercel's preview deployments, AI layer, and Secure Compute for enabling rapid growth and enterprise trust.", "body_md": "1,000+ legal requests managed daily across customer teams\n\n7-app turborepo monorepo deployed seamlessly to multiple Vercel projects\n\nMulti-step agentic legal workflows built end-to-end with AI SDK\n\nWhen in-house legal teams get a request from their business, it kicks off a manual process of pulling data from multiple systems. In a typical workflow, lawyers will pull deal details from Salesforce, review past contracts, and chase account teams for missing context, all before they start the legal work itself.\n\nSandstone automates this data gathering process, and centralizes requests into a single platform. When a request is created, Sandstone automatically ingests context from every system, including messaging and email tools. By the time the lawyer sees it, they have all of the information they need to get to work.\n\nThe founding team includes a McKinsey legal tech veteran, a lawyer-turned-engineer, and cybersecurity founder. At Sandstone, that mix shows up directly in the product: legal professionals work alongside engineers, catching the workflow details and legal nuances that only someone who has practiced law would notice. That is how Sandstone ships fast and correctly for one of the most demanding buyers in enterprise software.\n\nThat attention to detail has paid off: in less than 6 months after their launch, their revenue grew 40x. Three things made their growth possible on Vercel:\n\nPreview deployments fast enough to win enterprise trust\n\nAn AI layer flexible enough to ship agentic workflows in minutes\n\n[Secure Compute](https://vercel.com/docs/networking/secure-compute) that makes security a one-sentence answer\n\nFor Sandstone's customers, trust is built on responsiveness, and lawyers are trained to notice small details. Liam Germain, Co-Founder and CTO, puts it simply: when a customer asked for a scrollbar accessibility change on a Sunday, he had it live in production in five minutes. On any other platform, he said that same change would have taken five or ten times longer.\n\nVercel deployments help with more than just fixes. [Preview branches with inline commenting](https://vercel.com/docs/comments/using-comments) mean Sandstone's head of marketing can leave a note on a live landing page with Vercel Toolbar, which is configured to automatically creates a Linear issue and assigns it to engineering. What used to take an afternoon happens in a single comment.\n\n[Vercel Flags](https://vercel.com/docs/flags) runs feature flags across all applications, giving the team dynamic control over what is live, what is demoed, and what is still in progress, without touching a deployment.\n\nSandstone runs a [monorepo](https://vercel.com/docs/monorepos) of over seven applications across multiple Vercel projects with clean environment variable linking and a custom CLI built on top of Vercel's API layer, and every push on every repo generates a preview of production.\n\nSandstone's product started as smart triage, routing requests to the right lawyer. But the bigger opportunity was what happened when a request landed: every lawyer was still manually gathering context before they could act, pulling relevant details from a handful of separate systems, every time, for every request. Shipping a system that could handle that automatically required infrastructure capable of keeping pace with the product vision.\n\nSandstone has evolved into an AI system that executes complex, multi-step legal workflows end-to-end, searching through thousands of procurement agreements, surfacing the most at-risk agreements, alerting owners via Slack, and notifying legal when responses come in.\n\nVercel's [AI SDK](https://ai-sdk.dev/) is what makes those workflows composable enough for a small team to keep pace with a rapidly expanding product surface.\n\nThe composability shows up across the entire stack. Chat SDK powers the Teams integration. Emulate SDK handles integration testing for Slack and Okta. Agent Browser SDK runs agentic testing locally. Flags SDK drives the toolbar integration. One SDK, four workflows, zero separate infrastructure.\n\nSandstone uses Vercel [Secure Compute](https://vercel.com/docs/networking/secure-compute) to establish private communication between Vercel and its data layer. The setup was straightforward to implement and well-architected enough that it became a point of confidence rather than friction in every enterprise conversation. When the security question comes up with customers and prospects, and it always does, the answer is simple: all communication between Vercel and Sandstone's data layer is private and never touches the public internet.\n\nSandstone chooses platforms they believe will be ahead of the market in three to five years, a vendor evaluation framework they apply to both frontier model providers and core infrastructure.\n\nSandstone connected with the Vercel team through the Sequoia startup partnership. Within ten days of starting the company, the program gave them a shared Slack channel and access to engineers who could help think through architecture decisions in real time.\n\nLess than 6 months after launch, Sandstone had grown revenue 40x and landed customers across technology, manufacturing, and ecommerce, leading to a $30M Series A led by Lightspeed. The team is already shipping fast, so the goal for using that capital is straightforward: get Sandstone into more legal departments.\n\nThe vision they are building toward is a Legal Relationship Management platform that not only manages intake but also connects every workflow to the business context behind it, flags issues before they surface, and operationalizes legal knowledge through supervised agents. Sandstone describes this as a system of action for the legal department of the future.\n\nFor a team shipping with speed toward that ambition, the [infrastructure](https://vercel.com/blog/agentic-infrastructure) underneath the product has be a competitive advantage, not a burden, and that's why Sandstone runs on Vercel.\n\n**About ****Sandstone****: **Sandstone was born out of the acute frustration experienced by in-house legal teams who spend more time on process than progress due to fragmentation across disparate business systems. The platform provides a comprehensive solution for intake, triage, execution, fulfillment, and measurement of legal work, designed to run across email, messaging, and the business tools teams already use.", "url": "https://wpnews.pro/news/how-sandstone-grew-40x-in-147-days-on-vercel", "canonical_source": "https://vercel.com/blog/how-sandstone-grew-40x-in-147-days-on-vercel", "published_at": "2026-07-27 14:00:00+00:00", "updated_at": "2026-07-28 21:39:21.625620+00:00", "lang": "en", "topics": ["ai-agents", "developer-tools", "ai-products"], "entities": ["Sandstone", "Vercel", "Liam Germain", "AI SDK", "Secure Compute", "Vercel Toolbar", "Vercel Flags", "Linear"], "alternates": {"html": "https://wpnews.pro/news/how-sandstone-grew-40x-in-147-days-on-vercel", "markdown": "https://wpnews.pro/news/how-sandstone-grew-40x-in-147-days-on-vercel.md", "text": "https://wpnews.pro/news/how-sandstone-grew-40x-in-147-days-on-vercel.txt", "jsonld": "https://wpnews.pro/news/how-sandstone-grew-40x-in-147-days-on-vercel.jsonld"}}