"The ability to verify a business quickly, accurately, and at scale is no longer optional - it's mission-critical."
Plaid, Affirm, Bluevine, Pipe, Novo, Rippling — plus two of the three largest U.S. banks, 100+ regional banks and credit unions, and hundreds of fintechs
Instant access to Secretary of State records, IRS TIN verification, UCC lien data, and hundreds of government data sources. A business identity report in seconds.
Business risk scoring that captures fraud signals, adverse media, compliance flags, and entity legitimacy in a single integrated intelligence layer.
Ongoing surveillance for business changes - ownership updates, status changes, sanctions hits, and compliance events - across the customer lifecycle.
The hard part isn't any one of these things. It's connecting them. Ruppel built Middesk's architecture so that raw government data becomes structured signals, structured signals become risk intelligence, and risk intelligence becomes a programmable API that any developer can wire into their onboarding flow. The result: financial institutions that used to spend weeks on new business onboarding can now do it in under a minute.
The platform's latest evolution adds AI-powered web data enhancement - using a company's website and LinkedIn presence as identity signals - alongside automated TIN fallback logic and expanded adverse media coverage that mimics how a compliance analyst would search manually, but at machine speed.
In March 2025, Ruppel ran a webinar on the problem that keeps compliance teams awake: sole proprietors. They're 30 million American businesses. They often have no state-level filing. Many use their Social Security Number instead of an EIN. Some operate under fictitious business names with only local registrations.
They sit in what Ruppel calls a "gray zone between personal identity and business legitimacy." Traditional KYB tools treat them like incorporated businesses and fail. Traditional KYC tools treat them like individuals and miss the business context entirely.
His solution: build verification strategies that can operate at the intersection - pulling both consumer identity signals and business registry data, recognizing when a SSN doubles as a business identifier, and using web presence signals to fill the gaps where government databases run dry.
It's the kind of infrastructure problem that doesn't show up in a pitch deck but shapes what the platform can actually do at scale. Ruppel builds for the edge cases first. Everything else follows.
Ruppel's public presence is minimal and deliberately so. No long-form essays. No Twitter threads on startup philosophy. His personal site says he's working on "primitives and forward concepts" - and that's about it. The work speaks instead.
What colleagues and investors describe is a technical craftsman with strategic patience. He spent years at Zendesk and Checkr mastering what good engineering infrastructure looks like - not flashy, not overbuilt, but solid enough to serve millions of transactions without drama. That sensibility pervades Middesk's architecture.
The cognitive science background keeps surfacing in how he frames problems. Trust isn't binary. It's contextual. The same data point means different things to a bank, an insurer, and a marketplace. Middesk's platform was designed from the start to accommodate that variability - building one infrastructure layer that can serve radically different verification questions depending on who is asking and why.
Middesk is built for developers. Every product ships with a clean API so any engineering team can wire identity verification directly into their onboarding flow without building the data infrastructure from scratch.
He studied cognitive science at UC Berkeley - the science of how minds form knowledge and trust - before becoming an engineer. It shows in how Middesk frames verification as a trust problem, not just a data problem.
His Twitter handle @kurtruppel has been active since February 2009 - years before Middesk existed, and long before "founder Twitter" was a genre. Minimal output; clearly joined for the timeline, not the audience.
The name "Middesk" is a reference to middle office operations - the unglamorous risk, compliance, and operations layer between front-office sales and back-office accounting. Not a flashy name. That was the point.
Middesk entered Y Combinator's W19 batch and was among the select startups to arrive at Demo Day with a term sheet already signed. Accel and Sequoia couldn't wait for the presentation to finish.
The platform connects to hundreds of government agencies - Secretary of State offices, IRS, court systems, county recorders - to triangulate business identity from primary sources rather than aggregated databases.