Kristina Subbotina spent nearly a decade as a corporate startup lawyer, including a stint at Cooley. She left Big Law to become a VC investor, accidentally built a law firm instead, and has now converted that firm into an AI-powered legal platform called Lexsy that hit $372,000 in annual recurring revenue within six months of a $650,000 seed raise.
From Social Media to $1.3 Million
The path was indirect. Subbotina started posting legal tips for startups on social media to attract potential portfolio companies for a VC career. Founders kept asking for legal help instead of pitching deals. She launched Lawlace, a startup-focused law firm, and grew it to over $1.3 million in revenue in two years, driven almost entirely by social content that reached 5 million monthly views, according to Business Insider.
The AI Agent Pivot
By late 2025, Subbotina concluded that AI agents could handle the work of a junior associate. She raised $650,000 in seed funding with specific constraints: she wanted enough runway to pause sales for six months and build a productized version of the firm. “If an investor couldn’t help the business in other ways, such as media support or customer introductions, I didn’t take their money,” she told Business Insider.
The result is Lexsy, which Subbotina describes as an “AI-native company” with “an army of AI agents that take a first stab at everything while senior team members set strategy and review.” Clients buy a subscription and run their entire legal workflow on the platform, with attorney-client privilege covering all work. The human team consists of five people: Subbotina, a head of product and operations, and three software engineers, plus part-time support.
Revenue Conversion
Lexsy came out of stealth in June 2026 with $372,000 in ARR from three sources: clients migrated from Lawlace, word-of-mouth referrals, and Subbotina’s social media audience. That represents a 57% conversion rate from capital raised to annual revenue within a compressed six-month timeline.
Why Vertical Agents Keep Winning
The pattern is consistent across 2026’s agent startup landscape. The highest-velocity AI agent companies are not building general-purpose platforms. They are targeting professional services verticals where domain expertise, regulatory requirements, and high billing rates create natural moats. Taktile raised $110 million for financial decisioning agents. Probook raised $34 million for home services automation. Lexsy’s numbers are smaller, but the economics tell the same story: domain expertise combined with agent automation compresses the path from launch to revenue in ways that horizontal tools do not.