Lessons learned: How to get traction with AI and start building
Getting a grip on AI takes time, but after iterating through our first AI project, a method of comparing MARC records, we’ve gained enough skills and perspective to begin leveraging AI as a potential tool in our technology stack. There are many paths for incorporating AI, building a model from scratch, fine-tuning an existing model, or using a large language model through an API. However, we’ve learned through success and failure that there is a lot to consider before jumping into a solution. In this talk we’ll discuss how we’ve gotten up to speed, changed the way we collaborate, and the questions we now ask ourselves first when we consider using AI for a problem. We’ll talk about our current process for working with AI, and wrap up by sharing our ongoing AI work.