A legal cloud with a very narrow address
Dev Shrotri runs a software company that has spent a decade doing something most startups are told never to do. CloudLex, the New York firm he founded in 2015, sells to plaintiff personal injury law firms - and to almost no one else. There is no enterprise edition for corporate counsel, no general practice tier, no attempt to be everything for everyone. That restraint is the point.
Today Shrotri is the founder and CEO of a company of roughly 110 people that describes itself, plainly, as "the Legal Cloud." The product is a system of record for the messy lifecycle of an injury case: intake, matter management, documents, deadlines, medical records, settlement math, client communication. In January 2025 the company added Lexee, an AI assistant built specifically for personal injury work rather than bolted on from a general model. In July of that year CloudLex announced $3.7 million in fresh funding led by Aligned Partners, bringing its total to about $7.7 million.
What makes the story interesting is not the funding line. It is the conviction behind the narrowness. Shrotri has said the market gave trial lawyers two bad options, and CloudLex exists because he refused both.
Building after living the problem
Shrotri did not arrive at legal software as an outsider looking for a hot vertical. Before CloudLex he founded and ran Lexvia Inc., one of the larger civil litigation support services in the United States. Lexvia did the unglamorous work that keeps a personal injury practice moving: information management, remote paralegal support, full case management for plaintiff firms across the country.
That order matters. By the time he wrote a product spec, he had already spent years inside the operational pain he wanted to fix - the missed deadlines, the scattered records, the software that assumed every law firm was the same. CloudLex was designed by someone who had done the work first and built the tool second.
An engineer who chose lawyers
Shrotri is an engineer by training. He holds a master's in computer science and engineering and an MBA from Columbia Business School. His graduate research ran into multi-agent intelligent software systems, with projects tied to the EPA and DARPA, and he co-authored a book chapter on agent-based systems published through Carnegie Mellon and Springer Verlag. The subject - how independent software agents coordinate obligations and commitments - reads, in hindsight, like an odd rehearsal for coordinating the moving parts of a law firm.
His early career was in telecom and consulting. He worked as a software consultant at Sprint Communications in Kansas City, contributing to the Local Number Portability System, the plumbing that lets people keep their phone number when they switch carriers. He then consulted at Ernst & Young in New York, advising clients in telecom, media and networks. The through-line across all of it is coordination of complex, human systems - which is more or less what a personal injury case is.
The case for building narrow
The conventional startup instinct is to widen: more customer segments, bigger addressable market, a general platform. Shrotri went the other way and stayed there. Every feature in CloudLex assumes a plaintiff personal injury practice, which means the software can make choices a general tool never could - built-in statute-of-limitation tracking, settlement calculators, medical record retrieval, HIPAA-conscious storage, workflows shaped to how these firms actually run.
By 2022 the company was reporting around $12.7 million in revenue with a team near 80. The growth since then has been steady rather than explosive, which fits the founder more than a blitz would. This is a business built by an operations person who understands that in legal work, trust is earned slowly and lost fast.
AI without losing the human part
Lexee, the 2025 AI initiative, is the current frontier. The framing CloudLex uses around it is telling: automate the routine, but keep the empathy and human connection that personal injury practice depends on. Injury cases are, by definition, built on people at difficult moments in their lives. A product philosophy that treats automation as a tool for the lawyer rather than a replacement for them is consistent with everything Shrotri has built before - software that respects the work instead of flattening it.
The bet Shrotri is still running is a simple one, and unusually disciplined for a founder in a category chasing size. Pick one kind of firm. Learn it deeply. Build only for it. A decade in, the discipline has become the moat.