3 Startups · $24.5M+ · AI + Enterprise Solutions

What they're building

About 70% of how work actually gets done inside a large company is never written down. It lives in the memory of the veteran employee who knows the workaround, the right team to escalate to, and how the resolution plays out. When that person leaves, the knowledge walks out with them. Interloom builds a Context Graph: a continuously updated map of how operational experts actually resolve problems, built by ingesting millions of support emails, service tickets, call transcripts, and work orders. When a similar problem lands on a new employee or an AI agent, Interloom routes the resolution based on what actually worked before, not what a manual says. At Commerzbank, it reduced the gap between documented and actual operational knowledge from 50% to 5%. At Zurich Insurance, it beat 2,000 AI-native competitors in an internal AI competition for an underwriting use case.

Why this matters

Enterprise AI agents have a memory problem. You can give them the best reasoning model in the world and they still can't tell you how your specific underwriting team handles a complex broker submission, because that knowledge has never been written down anywhere. Interloom is solving the layer underneath the agent: the institutional memory it needs to actually be useful. The timing is backed by demographics as much as technology. The market is losing 10,000 baby boomers from the US workforce daily, and European enterprises face the same curve. Decades of operational knowledge is being permanently deleted at exactly the moment AI is expected to replace it. DN Capital's involvement carries a specific thesis signal: their lead investor Guy Ward Thomas was the first institutional backer of Cognigy, the German enterprise conversational AI platform, which DN Capital backed from its Series A and which was acquired by NICE in August 2025 for $955M, Europe's largest AI exit. He backed Interloom with that exact playbook in mind.