The Munich-based startup Interloom secured $16.5 million in funding.

      The Munich startup is developing what it terms a ‘context graph,’ a dynamically updated representation of how operational decisions are actually made within a company, based on millions of real cases rather than potentially unwritten documentation.

      There is a specific friction point encountered in every enterprise AI implementation, often faced early by anyone attempting to deploy an AI agent in a large organization. While the agent may have the technical ability to read documentation, follow instructions, and carry out tasks, it can't replicate the judgment of a person with fifteen years of experience who understands from firsthand knowledge why the standard procedures may not be effective on Tuesdays in the logistics department. This type of knowledge has typically not been documented because it was unnecessary to do so—until now.

      This is the issue that Interloom aims to address. The Munich-based startup, which has operations in Munich, Berlin, and London, announced on March 19 that it secured $16.5 million in a seed round led by DN Capital, with contributions from Bek Ventures and existing supporter Air Street Capital.

      This funding represents a significant increase from the company’s initial $3 million seed round, which Air Street led in March 2024 when the startup started revealing itself.

      Interloom’s primary offering is what it describes as a Context Graph: an ongoing model detailing how operational decisions are made within a specific organization, developed by gathering millions of real cases, support emails, service tickets, call transcripts, and work orders to identify patterns in how expert workers resolve issues.

      Founder and CEO Fabian Jakobi frames the challenge in terms of tacit knowledge, a concept introduced by British-Hungarian philosopher Michael Polanyi, who noted that most expertise cannot be fully expressed by the expert possessing it. Jakobi estimates that approximately 70% of operational decisions are never formally recorded.

      Jakobi likens this to Google Maps; just as the navigation tool learns optimal routes from real-time traffic data, Interloom maps out the routes that operational experts take to solve problems and uses this map to assist AI agents and new employees facing similar situations. The system is continuously updated, ensuring that every resolved case contributes to institutional knowledge rather than fading away when the person who handled it departs or retires.

      The potential retirement risk is part of the company's appeal. The press release highlights that around 10,000 baby boomers are leaving the US workforce each day, a statistic extensively reported by Pew Research.

      The underlying message is that companies are grappling with a growing challenge: the institutional knowledge accumulated over decades is vanishing precisely when AI is expected to automate complex operational tasks. Unless this knowledge is captured first, AI lacks valuable insights to utilize.

      Interloom’s initial customer base includes Zurich Insurance, JLL, and logistics company Fiege, along with Commerzbank and Volkswagen, the latter two verified independently by Fortune in its exclusive coverage of the funding. At Commerzbank, Interloom analyzed millions of customer support emails compared to internal documentation, reportedly narrowing the gap between documented processes and actual practices from around 50% to 5%.

      At Zurich Insurance, the company prevailed in an internal AI competition against 2,000 competing AI-native startups to secure a use case in underwriting, as described by Jakobi to Fortune.

      The investor roster supports its own rationale. Guy Ward Thomas, the DN Capital partner spearheading the investment, was the initial institutional supporter of Cognigy, the German enterprise conversational AI platform, which DN Capital backed from its Series A in 2019 and which was acquired by NICE in August 2025 for $955 million — recognized at the time as Europe's largest AI exit.

      Ward Thomas emphasized that the key takeaway from that investment was the importance of organizational context in effectively deploying AI agents. Mehmet Atici, who heads Bek Ventures' involvement, was an early investor in UiPath, the pioneering robotic process automation company that went public in New York in 2021. He argues that the current surge in AI agent adoption signifies the next significant shift in enterprise automation following RPA.