How to build context libraries and raise AI fluency
with Kene Anoliefo
Kene has led product and design at Google, Spotify, and Netflix. She founded Heard, an AI-powered user research platform, and now works on helping teams structure institutional knowledge for both people and AI. We talked about the three Ps of AI adoption — people, process, and platform — and why most organisations get the order wrong. Then we built context libraries live: structured repositories of strategy, design principles, and research standards that dramatically improve AI output quality.
What I took away
The framing of AI as "brilliant but clueless" is something I've since used with every team I speak to. It immediately reframes the bottleneck: it's not the model, it's the context you give it. Kene also made a point that's stayed with me — product leaders need to shift from being gatekeepers of expertise to architects of standards. That's an identity shift, not just a workflow change.
Where this connects
Building a context library for Atlassian's product team is something I'm now actively working on.