Spektr Raises $20M Series A for Compliance AI

Copenhagen-based spektr closed a $20 million Series A led by NEA, bringing total funding to $26 million. The startup builds configurable AI agents that automate manual KYC and KYB work for banks and fintechs, replacing hours of analyst research with minutes of automated investigation and structured risk rationales. Live customers include Pleo, Santander Leasing, Mercuryo, Phantom, and Monta. Founders are repeat entrepreneurs with a prior exit in identity verification. The new capital will expand spektr's platform, accelerate enterprise adoption, and scale integrations across compliance workflows.
What happened
spektr raised $20 million in a Series A round led by NEA, increasing its total capital to $26 million. The Copenhagen startup provides AI infrastructure that automates the manual work of compliance teams, specifically KYC (know your customer) and KYB (know your business) processes. Customers live on the platform include Pleo, Santander Leasing, Mercuryo, Phantom, and Monta. "Compliance technology has mostly focused on workflow and data collection," said Mikkel Skarnager, CEO and co-founder. "But the real bottleneck has always been the work itself... spektr automates those tasks with AI agents designed specifically for KYC and KYB compliance."
Technical details
spektr deploys networks of specialized AI agents that perform discrete analyst tasks: document review, ownership mapping, website verification, business-activity validation, and generation of structured risk rationales. The platform is configurable so institutions can design onboarding and monitoring flows, insert policy rules, and require human review gates at decision points. Key product capabilities include:
- •Automated corporate-structure and ownership mapping across registries and filings
- •Natural language extraction and synthesis of evidence into structured risk rationales
- •Configurable workflows that embed policy rules and human approval steps
- •Integrations with existing data providers and internal case-management systems
These agents are designed to complement, not replace, compliance analysts by producing auditable outputs that speed reviews and standardize decision documentation. The company emphasizes deployment control and governance, letting customers tune agent behavior to internal risk appetites and regulatory requirements.
Context and significance
Manual compliance remains one of the costliest operational functions in financial services. Years of investments in data aggregation and workflow tooling reduced friction but left the core investigative work to human analysts. spektr enters a growing category where applied generative AI is focused on automating specialized domain work rather than general chat. The team behind spektr are repeat founders with prior exits in identity verification, which reduces go-to-market risk for selling into regulated banks and fintechs. With NEA leading the round and participation from Northzone, Seedcamp, and PSV Tech, spektr now has both capital and investor network to pursue global expansion. For practitioners, this is an example of productizing AI agents around auditable pipelines and compliance governance, a pattern likely to repeat across other regulated verticals.
What to watch
Execution risks center on regulatory acceptance of AI-generated findings, integration complexity with legacy systems, and maintaining auditability for examinations. Watch for pilot outcomes at large banks, updates on governance features, and whether spektr extends into transaction monitoring or sanctions screening. If spektr demonstrates consistent reductions in analyst time and false negatives, it could become a template for industrializing domain-specific AI agents across regulated industries.
Scoring Rationale
This is a notable Series A for a startup applying generative AI to a high-cost, regulated problem. The funding and investor lineup raise the company's chance to scale, but the story is primarily operational rather than a foundational AI research advance.
Practice with real FinTech & Trading data
90 SQL & Python problems · 15 industry datasets
250 free problems · No credit card
See all FinTech & Trading problemsStep-by-step roadmaps from zero to job-ready — curated courses, salary data, and the exact learning order that gets you hired.