Transcript

You had this perfect data strategy and then the world changed. What was your earliest signal when you started realizing that the world was changing? So we had a a use case in which we were asked by the finance team, one of our earliest use cases was on our financial performance, specifically in software. The number one metric that you can measure is recurring revenue. So we would have this beautiful data around recurring revenue. We used it to to to do our end of quarter reporting etcetera. And we're going to make an agent around that. And so we built this beautiful agent, we deployed things on top of it. And it couldn't answer one question that they answered what they asked. It was it was actually a little bit embarrassing, like leave the most simple questions. And we started to realize that we were missing this translation layer. We, we had no way for us to interpret the human language against the structure of the data. And So what happened is we started to build it in this piece meal way to make these specific questions work. And then you have this realization like this could never scale. Like we're going to have thousands of of agents all over the place, each with their own custom built custom. Bespoke semantic layer and we really started to think about how we needed to think through the context for how people spoke to data and the importance of context and the importance of rules that help people understand what words and phrases mean. And that that was really kind of a wake up call and we've been focused on that ever since.