Brian O'Grady

for building a database.

This was the core requirement that our founders had when they started Quadrant.

They were trying to solve this exact problem of how do I do vector search at scale for e-commerce?

Theirs was actually a job matching website.

And they were getting choked on Elasticsearch because of the memory overhead of the JVM.

It just was kind of inhibiting them from taking full advantage of their resources.

They needed to kind of go in.

They needed to tune the JVM heap.

And they just felt that the functionality of, I mean, this was back in the day.

Elastic has added much more functionality since then.

This was back in like 2020 or 2021.

They were like, why don't we just build a system that is specifically designed to do this exact one thing?

So that's how Quadric came about.

And it fits in the stack as this dedicated engine that you use for vector search.

But the, where it becomes a bit more interesting is that, you know, you don't need to just use Quadrant to, let's say you're just, you think, okay, well, it's just for doing like e-commerce search.

It's actually not.

Quadrant is very flexible in how it can be deployed.

So I can deploy Quadrant in like a cluster mode and, you know, on like EC2 instances in my cloud.

I can deploy Quadrant via Docker locally.

I can actually deploy, and this is where I was taking a look at, Mike, your repositories.