Talky Just Raised $1.6M to Teach AI What Accountants Never Write Down

Jan de Vries · 2026-10-30

Most accounting knowledge never makes it into a manual. It lives in people's heads, in old spreadsheets, in the way one specific person always codes a certain invoice. Talky Labs thinks that's exactly the kind of knowledge AI should learn. And investors just bet $1.6 million on it. The Madrid-based startup closed a pre-Seed round in September 2026, pulling in about €1.5 million (roughly $1.6 million). The money goes toward team expansion, product development, and beefing up the technical infrastructure. ### Who's Backing Talky The round drew a mix of venture firms and angels. On the VC side: HelloWorld, Silence VC, Motive Force, and Aera Capital. The angel list includes Sebastián Gunningham, Remitly CEO Paul Golbaum, Lightbase CEO and former Seedtag CTO Javier Morenés, and others. For co-founder Maximilian Letón, the whole thing traces back to one night. "Two years ago, I met my co-founder Jorge Beamonte at a networking event," he said. "That night, I called my mom and told her: 'Mom, today is the most important day of my life. Everything is about to change.'" ### The Problem: Accounting Rules That Live Nowhere Here's the challenge Talky is chasing. Corporate finance teams make decisions using information scattered across ERPs, spreadsheets, emails, contracts, and people's memories. Standard automation tools handle the routine stuff fine. But unusual invoices? Exceptions? Those usually need a human to step in, which means delays and a higher risk of mistakes. Beamonte puts it plainly: "Every company has its own way of doing accounting, and it's almost never written down. Talky learns those criteria, applies them to the hard cases and turns them into information for better decisions." ### How the Platform Actually Works Instead of leaning only on predefined rules, Talky's AI studies a company's historical records and the habits of its finance team. It picks up how past transactions were handled, extracts the underlying logic, then applies that to new operations. When the system hits something unfamiliar, it doesn't guess. It flags the right employee to consult. What the platform can do: - Read invoices and suggest accounting entries - Review tax information - Reconcile bank transactions - Record results directly into the client's ERP system According to the company, the tech processes between 86% and 93% of accounting documents without any human intervention. Not bad for a team of nine. ### The Numbers So Far Founded in 2025 by Beamonte (CEO) and Letón (COO, 23), Talky reports it has already processed more than 1.8 million documents and over €2 billion in client spending through the platform. ### A Crowded, Well-Funded Space Talky isn't alone here. European investors poured money into AI accounting throughout 2026. Belgium's Dytto raised €1.5 million for its accountant assistant. Denmark's Plinto secured €2.9 million for automated financial controls. Germany's mika pulled in €6 million, and UK-based Zalos got €3.1 million for enterprise finance automation. Barcelona's Dolfin added €2.1 million for adjacent workflows. Add Talky's round, and you're looking at roughly €66.8 million across the sector. ### What's Next The plan now is to push into large enterprises and financial services providers handling high transaction volumes. Recruitment sits at the top of the founders' priority list. They want to grow the team while continuing to build out the platform alongside existing customers. For US finance teams watching European AI accounting tools, the takeaway is simple: the race to automate the messy, unwritten parts of accounting is heating up. And the startups that crack it could change how enterprises handle their books.