How Sandstone grew 40x in 147 days on Vercel
Sandstone pairs composable agent workflows with preview deployments and private data connectivity, showing the infrastructure a small team used to operate enterprise legal automation.
Sandstone manages **1,000+ legal requests daily** from a monorepo spanning **more than seven apps**. Its workflows gather context, analyze agreements, contact owners, and notify legal teams.
Agent builders should treat deployment previews, integration testing, feature flags, and private data access as parts of the product loop. Sandstone uses Vercel AI SDK components across Teams, Slack, Okta, browser testing, and toolbar workflows.
Sandstone manages **1,000+ legal requests daily** from a monorepo spanning **more than seven apps**. Its workflows gather context, analyze agreements, contact owners, and notify legal teams. Agent builders should treat deployment previews, integration testing, feature flags, and private data access as parts of the product loop. Sandstone uses Vercel AI SDK components across Teams, Slack, Okta, browser testing, and toolbar workflows. This is a vendor case study, and the **40x revenue growth in under six months** does not isolate Vercel or agent infrastructure as the cause. The material supplies no reliability, cost, or latency measurements.
This reinforces that production agents are integration-heavy products spanning identity, messaging, testing, rollout controls, and private data—not just model loops. It offers a useful deployment pattern across a multi-app monorepo, but the headline growth cannot validate the infrastructure choices, and missing reliability, latency, and cost data limit comparison or replication.