Amir built a model to automatically categorize and recommend scientific papers relevant to his research area. A small experiment - but the question it raised was bigger: what if AI could help experts navigate and build on knowledge at scale? That question never went away.

The idea sharpened through practice. Deploying NLP models at a major financial institution showed exactly where AI broke down in real organizations - not the algorithms, but the knowledge infrastructure around them. That same year, Amir started the AISC community to systematically study this gap with fellow practitioners.

Incorporated a company around a hypothesis already taking shape: the bottleneck wasn't AI capability - it was organizational knowledge architecture. Early work with teams confirmed this pattern across every vertical we touched.

The pandemic made the pattern unmistakable. Organizations that had systematized their expertise adapted quickly; those that hadn't lost critical knowledge when people left, priorities shifted, or teams went remote overnight. We watched it happen in real time across every industry we were embedded in.

Started building an AI-first knowledge management platform for corporate innovation teams. The pandemic had crystallized the problem; this was our first attempt to systematize the solution - turning institutional knowledge into something an organization could actually operate on.

Piloted the platform with corporate innovation teams. What we learned confirmed what we'd suspected: the hard problem wasn't technical. Organizations struggled not with the AI but with the underlying question of what knowledge to capture, how to structure it, and who owned it.

ChatGPT validated a decade of observation. Every organization suddenly faced the problem we'd been solving for years. We weren't pivoting - we were finally operating in a world that understood the question we'd been answering.

Formalized KnowledgeOps as a discipline. Began collaborating with economists and organizational theorists to codify what we'd observed. Published research with McGill University and presented at MODELS 2025.

Applied the framework to ourselves and to knowledge-intensive business processes for a handful of early clients. Our own operations: revenue grew 8x, team shrunk. For clients: faster knowledge transfer, reduced dependency on key people, and workflows that didn't break when things changed.

Taking the methodology to market at scale. The infrastructure is built, the proof is in our own numbers, and the demand is undeniable. The next chapter is helping every knowledge-intensive organization do what we did.