Munich-based AI startup Marble Cognition has raised $3.75 million in pre-seed funding to develop AI coworkers that learn how businesses operate by observing employees’ workflows. The round was co-led by Heartcore Capital and Seedcamp, with participation from angel investors including a board member of Datadog, a founder of Personio and executives from McKinsey, Mistral AI and Lovable.
Founded in late 2025 by Felix Rösner and Christophe Schmit, Marble develops AI agents that identify and automate repetitive administrative work. Rather than requiring companies to document their processes before introducing automation, the platform observes how employees perform tasks, creates structured workflow maps and deploys AI coworkers that execute those processes using existing business applications.
Building AI Coworkers That Learn From Employee Workflows
Many enterprise processes rely on knowledge that employees have developed through experience but never formally documented. According to Marble, approximately 90% of core processes at the companies it works with are recorded only in emails, spreadsheets or employees’ individual knowledge, making it difficult to introduce automation without first understanding how work is performed.
Marble addresses this problem through a desktop application that captures how employees interact with business software, including screen activity, clicks and keystrokes. The platform analyses these interactions to identify recurring workflows and convert them into structured process maps that AI agents can interpret and execute.
The company says its system anonymises captured information before it is accessed by people or AI models, including when processing takes place within a customer’s cloud environment. Monitoring is limited to work applications and excludes private applications, private browser tabs and activity outside working hours. Employees can pause recording, while collected information is analysed in aggregate rather than at the individual level.
Once a workflow has been mapped, Marble deploys AI coworkers that perform tasks across the tools employees already use. The system retains human involvement for decisions requiring professional judgment, allowing organisations to automate repetitive steps without transferring every decision to an AI agent.
At a private equity-owned financial services company in the US, Marble reports that one of its AI coworkers reduced handling time per case by 90%. During the first month, the team processed 50% more cases without additional hires. The AI coworker prepares settlement offers, which are then reviewed and approved by a specialist.
“Every company we talk to wants AI to take over real work. But nobody can tell the AI how that work is actually done,” said Felix Rösner, co-founder of Marble. “So we solve that first. Our Coworkers watch how the work happens before they take it over. They do it the way your best people do.”
Funding to Support Engineering and US Expansion
Marble currently works with technology unicorns, private equity-backed businesses and AI-focused companies in Europe and the US. Its customers include EGYM, Mutares and Enam, alongside other organisations seeking to automate administrative workflows.
The $3.75 million pre-seed investment will be used to scale the company’s AI inference infrastructure, expand its engineering team in Munich and launch in the US market.
Marble’s founders met at the Center for Digital Technology and Management (CDTM). Before launching the company, Felix Rösner worked at AI sales agent startup 11x, while Christophe Schmit, a former MIT researcher, worked at Enam, an AI startup backed by investor Elad Gil.
Reshma Sohoni, General Partner at Seedcamp, said:
“Most companies want to automate more, but the hard part has always been mapping out how the work gets done. Marble fills that gap, giving companies vital context for AI transformations that could define their future. We’re proud to back them from the start.”
Max Niederhofer, Partner at Heartcore, also highlighted the importance of understanding internal workflows before introducing automation. He described the ability to capture and structure company processes as a critical component of deploying AI agents in enterprise environments.
With the new funding, Marble plans to expand its technology and commercial presence while developing AI coworkers capable of automating business processes that have traditionally depended on undocumented employee knowledge.
