Skan AI’s Series C Bets Enterprise AI Needs a Map of Real Work

Skan AI, a Menlo Park startup whose software watches how employees actually move work through enterprise applications, has raised $63 million in Series C funding co-led by Cathay Innovation and Dell Technologies Capital, the company announced on August 12, 2026. Citi Ventures, Bloomberg Beta, State Farm Ventures, and Wipro Ventures also participated, taking total funding to roughly $120 million, and the round lands alongside the general availability of two products, Skan AI Blueprint and Skan AI Agents, that complete a three-part platform built on what the company calls a “context graph of work.”

The thesis Skan is selling is that enterprise AI agents fail in production because they are grounded in process documentation and system logs rather than in the work itself. Its software observes employee desktops across applications (the CRM, the email client, the mainframe session) and distills those observations into a continuously updated model of how a business actually runs. That model then grounds agents the company builds and operates for customers.

“Everyone is obsessed with building a better car. We think the bigger opportunity is building a better navigation system,” co-founder and CEO Avinash Misra said in the funding announcement.

What Skan AI Says Its Platform Is Already Doing

The company’s headline figures lean on banking and insurance deployments. At one top U.S. bank, Skan says it observed 11.2 million context switches across 1,500 finance professionals, uncovered $37 million in operational friction, and cut cost per transaction by 32 percent while lifting throughput 41 percent — producing $18 million in annualized savings. Cumulatively, the company claims more than $500 million in customer value, though Misra told VentureBeat that figure represents identified savings opportunities customers are working to recoup, not all banked savings.

On growth, Skan says revenue rose more than 300 percent year over year, for a second consecutive year, per VentureBeat’s interview, with average net dollar retention of 150 percent. It counts seven of the ten largest U.S. banks and a quarter of the Fortune 50 as customers, and says it has processed more than 25 billion work signals. The platform runs on NVIDIA AI Enterprise and NIM microservices, with Nvidia’s global head of banking quoted in the announcement on the appeal of agents that run on infrastructure a financial institution owns.

Two of the new investors came in through the customer list: State Farm Ventures and Citi Ventures were Skan customers before becoming backers, a detail Misra flagged in his own post about the round as the endorsement he values most. Cathay Innovation partner Simon Wu framed the bet as infrastructure: enterprise work context becoming “the foundational infrastructure layer for enterprise AI, the same way CRM became the system of record for customer relationships.”

The Capital Stack Behind the Round

The Series C is a reunion of existing backers rather than a changing of the guard. Cathay Innovation led Skan’s $14 million Series A, and Dell Technologies Capital led the $40 million Series B that added GSR Ventures and Liberty Global Ventures; both leads have now doubled down together. Citi Ventures has participated across all three rounds.

Misra and co-founder Manish Garg, childhood friends from Kanpur, India, and later classmates at the Indian Institute of Technology, founded the company in 2018 after their first venture was acquired by Genpact. The pitch has shifted with the market: from computer-vision process discovery, to process intelligence, to the context layer for agentic AI. The Gartner figures Skan cites for that framing — only 8 percent of enterprises have AI agents in production, and 95 percent of early implementations will require a complete redesign — come from Gartner research the company links in its announcement.

The observation-first model also carries an obvious tension, and Misra addressed it directly in the VentureBeat interview: Skan aggregates patterns across hundreds of workers rather than profiling individuals, scopes monitoring to opt-in applications and URLs, and keeps data inside the enterprise firewall, an architecture he says has passed review by European works councils. He acknowledged the technology has led some customers to cut headcount in certain processes, and in one anti-money-laundering operation he described, AI agents now run 60 percent of cases.