Brian O'Grady

I can deploy Quadrant on the edge to do vector search on the edge.

And we actually just released a demo that shows...

Quadrant operating on a pair of smart glasses to do image detection in real time.

So it's really cool.

And the key thing about Quadrant here is that it's not like just an embedded library, but it has the functionality to take the embeddings that you have locally on device, operate offline.

And then when you come back online, like let's say you're like a remote worker and you need access to like a database of like information.

you can access the database of this information locally using Quadrant.

And then if you're doing a chatbot and you're talking to a chatbot, for instance, and you want to then send your conversation history back to some remote database and do syncing, Quadrant can do this.

So it can sync a remote cluster with a local embedded device and manage the synchronization and the indexing of all the data.

So it's kind of this idea of

build once, deploy anywhere.

You can take Quadrant, you can put it on a pair of smart glasses, you can put it on your iPhone, you can put it on EC2 instance on your Mac.

You can even put it on a supercomputer.

There's been some research showing that Quadrant can run pretty efficiently on some supercomputers at the Argonne National Lab.

So,

Yeah.

So where I see a lot of people having success is that they start working with Quadrant on open source.

I think that from a developer experience perspective, it's very easy to work with.

People like it because if people are accustomed to, you know, standing up their stack in Docker or whatever, Quadrant is just like another few.

It's like a very simple Docker YAML.