Lunar founders raise €8.2m to launch AI-native audit startup Repodo

When the founders of a major European neobank decide their next move is auditing, you have to look past the boredom of the industry to see the logic of the trade. Ken Villum Klausen, the mind behind Lunar, isn't chasing a trend here; he is chasing one of the most inefficient, manual, and high-margin legacy sectors left in the corporate world. His new venture, Repodo, just pulled in 8.2 million euros in pre-seed funding, which is a massive number for a company that is essentially still in the garage phase.

The Audit Bottleneck

If you have ever built a company, you know the audit cycle is where productivity goes to die. It is a months-long grind of gathering documents, explaining spreadsheets to junior associates, and paying six-figure fees for what amounts to a checklist. The current system relies on the Big Four firms throwing thousands of human hours at problems that are essentially data reconciliation tasks.

Repodo is betting that the Large Language Model era makes the old way of auditing obsolete. The goal isn't just to make a better spreadsheet; it is to build an AI-native system that understands the context of financial data. For a builder, this is the ultimate use case for AI: taking a high-trust, high-friction process and automating the validation layer.

Why Neobank Founders?

It is worth asking why the Lunar team is doing this. Neobanking was about the interface and the user experience, but it was also a masterclass in dealing with regulators. Klausen and his team spent years navigating the strict compliance requirements of the European banking sector. They didn't just see the front end; they saw how broken the back-office reporting systems were.

This pre-seed round, led by EQT Ventures, suggests that investors are no longer looking for general-purpose AI. They are looking for founders who have actually lived through a specific pain point and have the technical depth to automate it. In the case of Repodo, the pain point is the annual audit, a mandatory requirement that every significant business in the world has to endure.

The Tech Stack Reality Check

We need to be honest about the difficulty here. Building an AI that can "read" an invoice is easy. Building a system that a regulator or a partner at an accounting firm will trust is an entirely different mountain. The risk in this sector isn't the technology failing; it is the human gatekeepers refusing to accept the results.

Repodo is positioning itself as an AI-native firm, which usually means they are building their workflows around the LLM rather than just plugging an API into an existing dashboard. For developers in the crypto and AI space, this is a signal. The money is moving away from the infrastructure layer and toward the vertical application layer. We have enough models; now we need the agents that can actually perform a job functions.

What This Means for Founders

If you are building in the AI space right now, Repodo’s raise should tell you two things. First, the size of pre-seed rounds for proven founders is still healthy, despite the broader market cooling. Second, compliance and "boring" B2B services are the gold mines of this decade.

Most AI startups are trying to build creative tools or chatbots. Very few are trying to solve the legal and financial verification problems that keep CEOs up at night. If Repodo can prove that an AI agent can perform a standard audit with 99% accuracy, they won't just be a software company; they will be the new infrastructure for the global economy.

The Skeptic's View

I am always wary of the "AI-native" label. It is often used to mask a lack of proprietary technology. However, when you have a team that has already built a regulated financial institution, you give them more leeway. The challenge for Repodo will be the data silo problem. Companies are notoriously protective of their financial data, and getting them to feed that data into a startup's AI model—even one led by banking veterans—is a significant hurdle.

Furthermore, the audit industry is protected by heavy regulation. In many jurisdictions, a human must sign off on the final report. Repodo isn't just fighting technical problems; they are fighting a century of professional tradition. They will likely start as a tool for auditors to increase their own margins before they ever become a replacement for the auditor themselves.

Building for the Long Game

The 8.2 million euros isn't just for coding; it’s for the long, expensive process of gaining trust. For builders, the lesson here is that if you want to disrupt a legacy industry, you need a war chest and a deep understanding of the regulatory landscape. You can't just move fast and break things when you are dealing with a company's financial integrity.

Repodo is a company to watch, not because auditing is exciting, but because it represents the first real attempt to use AI to dismantle one of the last bastions of the professional services elite. If they succeed, the "Big Four" might soon find themselves competing with a lean team of engineers and a cluster of GPUs.

Takeaway

The era of generic AI is over. The next winners will be the founders who take AI into the most regulated, manual, and expensive corners of the business world. Auditing is the perfect target.

Read the original at Sifted →