Subscribe
Share
Share
Embed
How AI is Built dives into the different building blocks necessary to develop AI applications: how they work, how you can get started, and how you can master them. Build on the breakthroughs of others. Follow along, as Nicolay learns from the best data engineers, ML engineers, solution architects, and tech founders.
Nicolay (00:00.95)
Hey buddy, today we are launching the next season for How AI is built. The last few weeks we spoke a lot to regular listeners, but also past guests on how to improve the podcast. We also looked at our episode data, analyzed it, and we will be applying a lot of the learnings we found and a lot of the feedback we got to season two. And this season will be all about search. And we will be trying to make better, more actionable
and more in -depth episodes. And the goal is that at the end of this season, you will have a full -fledged course on search, on all things search, with a bunch of mini courses on specific areas, like for example, RAC. And we will be talking to experts from different fields, like information retrieval, information architecture, recommendation systems.
from academia and from industry, fields that usually don't talk to each other. And we will try to unify and transfer the knowledge to give you a full tour of search so you can build the next application or the next feature you're building that has a search component with more confidence. And you actually know what technologies might be suited here, what technologies wanted to apply, and how to apply it.
For example, we will be talking to Charlie Hull on how to systematically improve your search system. We will be talking to Nils Reimers on the fundamental flaws of embeddings and how to fix them, with Daniel Tunkelang on how to actually understand the queries of the user, and many more people. We will try to bridge the gaps between the different areas. So for example, how to use the decades of research and practice
on iteratively improving traditional search system like Elastic and apply it to RAG How to take new methods from recommendation systems and vector databases and bring it into traditional search systems and how to use all of the different methods even together as search signals and combine them to deliver the results your user actually wants. And for that,
Nicolay (02:24.342)
we will be using two different types of episodes. One, traditional deep dives, like we have done them so far, and each one will dive into one specific topic within search, interviewing an expert on that topic. And the second type will be supplementary episodes, which answer an additional question, often that's complementary to the episode of the week, or which handle some...
cursory knowledge for the episodes, which we did not get to in the deep dive. And we will be starting with episode next week's looking at in the end, the first, the last, and the overarching action in search, actually understanding the user intent and understanding the queries. And this will be a really exciting episodes with Daniel Tunkelang. And I'm really excited to kick this off. What I would love to hear from you.
Like first, what would you love to learn in this season? Second, what guests should I also have on? And thirdly, like what topics should I make a deep dive or a supplementary episode on? And especially if in one episode a question pops up or something is unclear, it's likely that you're not the only one. So just shoot me message on LinkedIn, on Twitter, dive into the DMs or let me know in the comments.
And I will try to answer it either in the comments or do an episode to answer it for everyone. And if I think more people might have some uncertainty about that area as well. And yeah, I think enough of me rambling. Let's kick it off and I will see you next Thursday when we will start with query understanding. See you.