From My Workflow to Ours: What We Learned About Scaling AI Across Data Teams

​Your data team started using AI - shipping code and publishing analyses faster than ever. You expect real impact on the product and the business.

​But it doesn't quite happen that way.

​Everyone runs their own workflow locally (or worse, five different scripts doing the same thing). It's hard to trust the data AI provides. Every new hire re-teaches the model how the business actually works. One schema change triggers a chain of errors - and no one's quite sure who's responsible for maintaining any of it.

​Adi Masad will share how her team at Honeybook built a shared Data & AI infrastructure - one that actually creates and maintains itself (well, almost). She'll walk through how to approach a project like this even in a small-to-mid-size startup, and what it really takes to change how an entire team works.

​What worked, what didn't, and why it pays to learn from other people's mistakes.