Brian Maucere

It said, in its latest form, our monitoring systems reviewed interactions within 30 minutes of completion, categorizing the interaction and assigning it a severity level.

As the latency decreases towards near real-time review, the security benefits increase significantly.

Eventually, the monitor may be able to help evaluate coding agent actions before they're taken.

I think that's so critical to point out.

Providing another important defense in depth control alongside other existing security monitors.

So just think about what they're talking about there and saying like, right now we can do this in 30 minutes post coding.

We can use our other smaller agents to go in and help.

But as this moves towards real time,

it'll be catching errors before the coding is ever written to begin with.

And I think this is the path forward, right?

This is what everybody wants to hear is like, hey, you know, if I'm going to move from something that's, you know, inherently considered vibe coding, whether that's a dirty word or not, depends on the business of whether they feel like vibe coding isn't real coding or whatever, right?

It's not real code.

It's going to break on you instantly.

It doesn't have the backend.

I've heard it all.

And the reality is maybe some of that is true and maybe I wouldn't stand up full enterprise solutions on Amazon.

something that came out of clogged code, but increasingly so whether you're using codex or clogged code, or now we'll talk about it a bit, uh, Google, you know, AI platform who can do a lot of the same things now with builder, um, these problems are going to go away.

And so the idea that you can't trust the code, I think is going to be a 2026 problem, not a 2027 problem.

Yeah, I think it's a good point.

Look, I'll still roll back to what I just said.