New Course: Get AI-Ready With Erik!
Video Summary
In this video, I introduce a new course titled “Get AI Ready with Erik,” designed to help you navigate the upcoming AI features in SQL Server 2025 and beyond. Given that Microsoft is positioning SQL Server as AI-ready, it’s crucial for professionals like us to be prepared to work with these new capabilities. The course focuses on practical applications rather than deep theoretical knowledge, teaching you how to leverage vector search effectively without getting bogged down by the complexities of underlying mechanisms. We cover everything from foundational concepts to advanced topics such as hybrid search and multi-vector search, ensuring that you can make informed decisions about when and how to implement these features in your production environment.
Full Transcript
It’s Friday and we’re introducing a new course. It’s called Get AI Ready with Erik because Microsoft says SQL Server 2025 is AI ready and you need to be ready to work with AI in SQL Server. Whether you care about it or not, whether you think it’s good or bad or not, at some point you’re going to have to deal with these new features and I want you to do it in the least stupid way possible. The whole foundation of the course is based on this premise. You don’t need to know how it works. You don’t need to know all the crazy math. You don’t need to know all the weird theory. You don’t need to know all the, you know, gears and whistles and bells and stuff that gives people six fingers on the, on the backside. You just need to know how to work with these features in smart ways. That’s what I want to teach you. The whole thing started because I do SQL Server performance tuning and I saw this whole thing, the whole suite of new things hit SQL Server and I said, gosh, the way things are going, I’m going to be a little bit more. I’m probably going to have to do some performance tuning work on this. And it turned out on my path to learning how to do performance tuning work on this. I had to learn a whole lot about how all the stuff worked inside. So that’s my distillation here, how all this stuff works. Now you can work with it too. Vector search is fairly easy to grasp as a concept. It stores these embeddings, which are a series of floats, which might, I guess, store it as a JSON. And then when you change other text into a similar, embedding another array of numbers, vector search can tell you how close, how similar those strings are without a bunch of nasty wildcard searches and stuff, right? No longer just like saying where something is like, you know, percent word, percent word, percent word, or like, you know, just long strings of or clauses doing this stuff. It allows you to sort of search by meaning and intent, not just like, you know, normal keyword searches. So you can find similar things pretty fast, even on pretty big tables.
It’s not perfect though. Details often get lost. And sometimes you still need to do some keyword searching and sometimes you still need quality filters on things to make sure that, you know, not only is the content that you’re providing in your search is relevant, but also that it’s not just like garbage, right? So, you know, that’s no fun. Microsoft is, you know, Microsoft is, you know, of course gets a late pass on this one. That’s, you know, behind the curve a bit on this. Most other, you know, of their competing databases have had this stuff around for a while, but, you know, moving data to another database, as I’ve learned from seeing so many people talk about how they’ll be in Postgres in three to six months and they never seem to get there. Your data is already in SQL Server, moving it off to either a specialized database or a whole other data platform just to take care of vector search requirements is probably a non-starter for a lot of people.
You’ve got everything already in SQL Server. Adding a vector column or a table with a vector column in it beats migrating for most people in the world. Microsoft was late, but, you know, they actually showed up. You have some generally available features available right now and you have a lot more features still in preview.
Will they ever be real? I don’t know. Will Microsoft ever finish anything again in SQL Server? I have no clue, but I can tell you what’s there and what’s not. So this is what the course covers. It’s in eight parts over around 28 videos. It’s about 10, 12 hours of total content.
We cover foundations, practical applications, hybrid search, vector indexes, native integration, managing embeddings, production patterns for dealing with all this new vector stuff. And then some more advanced topics like rag and multi-vector search. It’s a pretty solid course, I think.
And one thing that I want to say here is that, you know, where a lot of this stuff is in preview and like maybe isn’t out yet in SQL Server 2025 is RTM. Like we don’t even have one cumulative update for it as of this recording. I’m going to be adding in stuff as cumulative updates add, you know, substantive changes and new stuff to SQL Server.
So this isn’t like a course that’s just going to rot on the vine. As Microsoft works on this, the course will get updated alongside it. So there will be new content to cover new things as the features move along.
What I hope you get out of this is understanding when vector search makes sense. Because when your boss comes along and says something like, hey, we want to do this and here’s what we want to do it for, you can speak intelligently about it. You have some background.
You have some knowledge. You have information. You can feel empowered to speak authoritatively on the subject. Right. We want to know what’s production ready versus what’s still in preview today, how to monitor and troubleshoot performance issues and sort of like capacity planning and all the other stuff that goes along with, you know, DBA and developer tasks. And we want to know how we can implement this in production and make working with the current set of limitations as pain free as possible.
So right now, if you go to training.erikdarling.com, there should be a link down in the video description directly to this course as well. You can use the coupon code AIREADY to knock $100 off the cost of the course. That coupon code is available for the first 100 people who sign up.
So if you want to be one of the lucky hundred to save $100, I suggest you get moving because things are already flying off the shelves and, you know, I would hate to see you miss out on this wonderful opportunity. Anyway, thank you for watching. I hope you enjoy the course and I will see you, well, when I see you.
All right. Goodbye.
Going Further
If this is the kind of SQL Server stuff you love learning about, you’ll love my training. Blog readers get 25% off the Everything Bundle — over 100 hours of performance tuning content. Need hands-on help? I offer consulting engagements from targeted investigations to ongoing retainers. Want a quick sanity check before committing to a full engagement? Schedule a call — no commitment required.