Julian Rowlands spent years in the trenches of startup finance—CFO at Indonesia's Xendit during a pivotal Series B, then leading finance at Spruce before its acquisition by Zillow in 2023. He knows the frustration of wrestling data from a dozen disconnected systems, each with its own logic, its own dialect. And he's betting that frustration is about to become an opportunity.
Cashboard, the New York startup Rowlands founded, just closed a $5 million seed round led by FINTOP Capital, the firm announced July 22, 2026 via Axios Pro. The company emerged from Y Combinator's Summer 2022 batch and had previously raised $1.5 million in pre-seed funding from TTV Capital back in July 2022. Now it's trying to carve out space in what has become, perhaps unexpectedly, one of the year's more crowded enterprise software niches: the semantic layer for finance.
The pitch sounds technical because it is. Cashboard bills itself as infrastructure that sits between finance data and AI—a translation layer, essentially, that lets large language models like Claude function as what Rowlands calls a "trusted FP&A machine." The platform taps into more than 900 integrations: payroll systems like ADP, CRMs including Salesforce and HubSpot, accounting tools such as Ramp, data warehouses like Snowflake, planning software from vendors like Workday Adaptive Planning. It's a sprawl that mirrors the reality inside most finance departments, where data lives everywhere and nowhere all at once.
But according to Cashboard's marketing, the real sell isn't breadth of connections—it's control. Finance teams use Cashboard to build what amounts to a data dictionary, complete with standardized metrics and granular permissions that drill down to specific rows and columns. Once that governed layer exists, it gets exposed to both AI agents and human analysts. The company offers an MCP (Model Context Protocol) integration for Claude, scheduled AI agents that generate recurring reports without prompting, and an Excel plugin for one-click data retrieval. It's governance by design, a response to the chaotic early experiments many companies had with AI poking around sensitive financial datasets.
The timing feels deliberate, if not inevitable. The first half of 2026 has seen a wave of announcements in this corner of enterprise tech. OneStream, a more established player, rolled out what it describes as an "agentic" finance layer. Incorta partnered with Workday on semantic planning data. Startups like Astrada and FinBoard raised funding with strikingly similar positioning—governed data layers, AI-ready finance infrastructure, semantic models designed for the Office of the CFO. An AtScale report from January flagged finance as a key early adopter of semantic-layer technology, which tracks with what venture investors have been saying privately for months.
Whether the category can support this many entrants remains an open question.
Rowlands, for his part, seems intent on building credibility beyond the product itself. In mid-2026 he launched a "CFOs on AI" interview series featuring finance leaders from OpenAI and FinOptimal—a content play, sure, but also a signal that Cashboard wants to be part of the conversation at the CFO level, not just the data engineering team. The company holds SOC 2 Type II certification, maintains offices in New York with additional presence in Bellevue and Toronto, and shows roughly 11 to 50 employees on its LinkedIn company page as of July 2026, though exact headcount has not been independently verified. Recent job postings reference a "Founding Onboarding & CS Lead" role, suggesting the company is gearing up to scale deployments.
Cashboard markets a six-week onboarding timeline that includes free professional services to model data sources, migrate mappings, and go live. That kind of white-glove treatment doesn't come cheap, which makes the $5 million seed raise—modest by recent fintech standards—all the more interesting. The company hasn't disclosed how it plans to allocate the capital, but the emphasis on onboarding and customer success hints at a focus on getting early customers into production rather than splashy growth metrics.
For finance teams drowning in spreadsheets and stitched-together workflows, the promise of a single governed layer that speaks to AI is compelling. Whether Cashboard can execute on that promise—and differentiate itself in an increasingly noisy market—will determine if this is the start of something durable or just another well-funded experiment in enterprise infrastructure. Rowlands has done the CFO job. Now he gets to sell to the people who used to be his peers.
