James Sturges

How do you curate knowledge?

How do you find stale knowledge?

You know, things like that, answer employee questions.

And so a lot of our product has sort of evolved around how do we do that with machine learning and do this much smarter?

So how do you give access to knowledge

to the enterprise in a much better way and give someone a better experience than browsing through thousands of levels of folders in folder structure or in things like SharePoint or other things like that and instead take a natural language query and ask a question and basically get an answer.

And not only get an answer, but get here's the authoritative answer, here's some related things, and here's some other similar things you might want like documents.

So the knowledge base is, we do two things.

One is a lot of companies already have this information.

And so we don't want them to do a lot of work to bring it over to our side necessarily.

So we'll go read a whole bunch of documents from file shares or other systems they may have, intranets, things like that.

And then basically learn what type of information it is, what kinds of knowledge can this particular piece of content answer.

What's that?

No, no, no.

This is all using machine learning, yeah.

Yeah, so we're really focused on the enterprise.

So it's not so much... Google Drive is something we could support.

Most of our companies have things like FileShare, SharePoint, Igloo, Intranets.

These are things that have hundreds to thousands or more of documents.

And sometimes that also is other disparate systems like...