Sandstone has raised $30 million because investors see a large legal market that AI has not yet properly reached: the in-house team.
Legal AI has been one of the loudest corners of the startup market, but most of that noise has been aimed at law firms. Sandstone is taking the other side of the legal world, the corporate legal department, and its new Series A suggests investors believe that is where a different kind of value will be created.
The company announced a $30 million Series A on June 9, led by Lightspeed Venture Partners. Existing backers including Sequoia, Mantis VC, SV Angel, Operator Partners, Kearny Jackson, Daybreak Ventures and Litquidity Ventures also participated, according to a report from TechCrunch. The round comes only six months after Sandstone launched publicly with a $10 million seed round led by Sequoia in January.
That speed matters. Legal technology has often moved slowly because lawyers are cautious buyers, risk is hard to abstract away, and workflows are full of exceptions. AI has changed the investor calculation. When repetitive professional work can be routed, drafted, searched, measured and improved by software, a market that once looked resistant to automation starts looking like an obvious target.
Sandstone is not trying to be an AI lawyer in the same sense as Harvey or Legora. Those companies have become better known for serving private practice, where lawyers need help with research, drafting and legal reasoning. Sandstone is focused on the operational mess inside companies: Slack requests, email threads, Jira tickets, contract reviews, business questions, scattered playbooks and the constant movement of work between legal and the rest of the business.
That distinction is important because in-house counsel are not paid by the hour. A law firm can admire efficiency and still live inside a business model where time is the unit being sold. A corporate legal team is measured differently. It has to reduce bottlenecks, protect the company, support sales and procurement, and keep business teams from waiting too long for an answer. If AI can remove friction there, the return is easier to understand.
Sandstone describes its product as a legal context engine and operating layer for in-house teams. Its site says the platform connects requests, workflows, playbooks, documents, business context and reporting across tools companies already use, including email, Slack, Microsoft Teams and other systems. The promise is not just faster document review. It is a more organized legal function, where knowledge is captured as work happens and reused instead of being trapped in old files or a senior lawyer's memory.
That is the kind of problem that sounds boring until you have lived inside it. A sales team wants a contract approved. Procurement needs terms checked. Finance wants risk visibility. A business stakeholder asks the same question legal answered three weeks earlier. None of this is glamorous, but it is exactly where corporate teams lose time and trust.
Why Investors Are Moving Faster
The six-month jump from seed to Series A is the most telling part of the story. It shows that venture investors are applying the AI infrastructure playbook to vertical software: find a huge knowledge-work market, build deeply around workflow, then move quickly before the category hardens around a few winners.
Sequoia's earlier investment note on Sandstone framed law as a strong use case for large language models because the profession runs on text, judgment and repetitive high-cost work. That is true, but the more practical point is that legal teams have rich internal context. Contracts, policies, prior decisions, risk tolerances and negotiation patterns all matter. Generic AI tools struggle when they do not know how a company actually works.
This is where specialized vertical AI has an advantage. A tool built for everyone has to stay broad. A tool built for in-house legal can understand intake, triage, permissions, approval paths, clause libraries, playbooks and reporting. It can also meet lawyers where they are, which is usually inside the tools the rest of the company already uses.
Sandstone will still face serious pressure. Frontier model companies are moving into legal workflows, and Anthropic has been expanding Claude for Legal with features for case law search and deposition preparation. Large enterprise software vendors will not ignore legal operations either. The question is whether a startup built around the specific habits of in-house counsel can stay ahead of broader platforms that have more distribution.
The answer will depend on trust as much as technology. Legal departments handle sensitive material, and Sandstone's own security materials emphasize controls such as audit logs, role-based access, SAML single sign-on, data portability and a contractual guarantee that customer data is not used for model training. Those details are not decoration. In legal AI, security is part of the product.
The bigger market implication is straightforward. AI is moving from chat interfaces into operating systems for professional work. In-house legal teams are a natural place for that shift because the work is expensive, repetitive, context-heavy and connected to revenue. If Sandstone can turn legal from a request queue into a better coordinated business function, this round will look less like hype and more like an early marker of where enterprise AI is headed next.
Also read: Bitcoin is not dead, but the easy crypto story is breaking • Standard Bots becomes a robotics unicorn as US factories automate • Lovable’s $500 million run rate makes vibe coding harder to dismiss
