In this episode of Just Curious, host Stu Willson sits down with Felix Rösner, co-founder of Marble, the AI platform that builds “AI clones” of organizations by capturing thousands of hours of real work. Felix explains why most automation efforts fail, why context—not tools—is the real moat, and how Marble helps PE-backed companies, roll-ups, and complex service organizations uncover hidden inefficiencies before automating anything.

Chapters:

00:00 – Intro
01:30 – Felix Rösner’s background and the origins of Marble
03:45 – Why “context is the last moat” in AI
05:20 – What it means to instrument and model real workflows
07:00 – The insight that sparked Marble's approach
09:10 – Why manual shadowing fails and automation efforts break
10:30 – What Marble’s “AI clones” actually are
12:20 – Balancing workflow visibility with privacy and compliance
14:10 – Signs your org doesn’t understand its operations
15:30 – Which companies feel the pain most acutely
17:10 – Case Study: PE-backed roll-up struggling with workflow clarity
19:00 – Deploying agents and capturing 10,000+ hours of work
21:20 – How employees respond to workflow capture
23:00 – What Marble does with the data it records
26:40 – Key insights surfaced for leadership
28:30 – The 40% redundancy: where time actually went
34:00 – What kinds of repetitive tasks were automated
36:50 – Advice for COOs and operations leaders
38:20 – Closing thoughts

📩 Subscribe for Weekly AI Insights & Case Studies:

https://just-curious-ai.beehiiv.com/?utm_source=home&utm_medium=video&utm_campaign=FELIX_RÖSNER_INTERVIEW

‍