Michael Dominick

What do we mean by that?

What does Thousand Eyes mean by that in particular?

So it's really interesting.

So let me just kind of recap that.

So the AI was able to see that something was going wrong in the loop, that there was a latency spike, and then send that information to the LLM, who then was able to send it to an MCP to basically diagnose the problem?

Right.

Interesting.

So it's acting as almost a, I guess, a bridge or a negotiator between the different AI systems.

Right.

Because theoretically, if you just adhere to the standard tightly, you could write an MCP for almost anything.

I mean, lots of things already have MCPs.

You mentioned GitHub.

I think Atlassian does too, right?

Yeah.

Yeah.

Okay.

So your internal systems, let's say you have an internal enterprise application.

I don't know, like some kind of ERP or something.

You theoretically could have that hooked into an MCP that also communicates with ThousandEyes to...

I mean, maybe I'm getting a little too paranoid in my work with the MRO people, but look for weird network incursions, right?