When our nurses and doctors are sitting down with a patient, they may need to grab certain information about a drug or a procedure. They need access to that documentation instantly, And they've always relied on my team to provide a solid search solution for that. I'm guessing it's typical for all kind of large bodies of content, but we have eighty percent of our users looking for twenty percent of our content. The other twenty percent of our users are looking for these random things that are spread across this other eighty percent of content. So you get the the kind of long tail effect where you can't predict what someone's going to be searching for, can't predict how they're going to search for it. And we spent so much time over the years trying to play the game of manipulating the search rankings and manipulating featured results and adding synonyms and trying to always stay one step ahead of our users, machine learning has done away with that. Now the machine learning is doing all that heavy lifting and we sit back and get to watch and celebrate a successful platform. So we're an academic medical center, not for profit. Do about three billion a year in revenue with about six hundred thousand patients in South Central Wisconsin. Our goal at UW Health is to provide remarkable patient care for patients and families. Part of providing that care is getting them the right information at the right time. If a physician or a nurse can't find the document they need, can't find the information they need, Now our patient experiences suffering because they're waiting. They don't want to wait. If heaven forbid they find the wrong information, that could be even worse. My team is comprised mainly of, developers and systems engineers. We build custom web applications that serve various, enterprise needs. We oversee all the marketing technology stack. We, develop and maintain our hospital intranet. All of the clinical knowledge management passes through my team. We also provide all of our search capabilities as well for both public facing and internal websites. We hear every day from end users how critical it is With past experiences when we went live with a new search product, it was never smooth. There was always dramatic changes in the search results and then it was weeks and months of trying to Fine tune that search experience to get it back to where it needed to be. With Kaveo, we went live. We turned the Google search appliance off and turned the Coveo on And we got no response. So to put that into context, previously we got a flood of support requests and flood of angry emails saying what's wrong with the search. Why isn't search working? On day one of Coveo, typed in their normal search term and there it was there. That was without machine learning turned on because we hadn't we didn't have any data yet. But then a month, two months later when we had machine learning Actively working. We saw our results to get even better. Eat Better. Even better. And it further solidified the fact that we no longer have to play the game of trying to keep up with our users and anticipate what they might search. Machine learning model takes care of that. Really, our goal is that search and all the technology behind it is kind of an afterthought for the end user. They don't need to worry about how it works. They just know that it works and that they get what they need.