Oct 6, 2026 · 17 min · 8 segments
What if your next troubleshooting session started with a conversation instead of digging through logs? Recorded at VMware Explore 2026, Pete Flecha and John Nicholson welcome Victoria and Dave…
Victoria GimmaGuest
John NicholsonHost
Pete FlechaHost
David LinthicumGuest
So one of the things I think people are confused about when they hear about agentic, they think it's a replacement thing.
The really thing that I'm finding as I'm setting up these systems is it turns your agents into a true force multiplier for the people.
I would also add on to that by saying what the VMware product engineering team has done here in terms of a very intentional way to execute this from a very safe lens, right, is making sure that this feature is completely compliant with any customer's kind of compliance regulations, right? Having it completely anchored in the data center, having it completely air-gapped if they need it to be, and self-efficient in terms of everything being located on-prem, but more importantly around the agent isn't going to be able to be this kind of toggled overachiever.
They're going to hunt that to the end, even if they have to blow up the data center or hack a frontier lab to accomplish it.
I mean, it's like you ask an agent today, outside of what we're doing, to go and execute a task.
if there's a piece of the task that they're not able to execute because they don't have line of sight or they don't have access, they're going to elevate permissions and they're going to figure out how to get that access to execute the end goal, which is the ask from the user.
And so these are all the areas in which our product engineering team has been very forward thinking around and ensuring does not happen.
So what they have really done in terms of nesting in existing, like, you know, carefully crafted skill sets for these agents, giving them the right level of constraints and guardrails so that even the IT end user can be very confident and they can trust this agent once they ask it a question and have like this, you know, long conversation around, you know, whatever it is they're trying to achieve, they'll always know and see what the agent is doing because there's that level of transparency and what it's executing, where it's looking for the information and doing the next steps.
But then also it'll be very confident in the follow through and know that if it needs to elevate permissions, well, that's a human level intervention that is very intentional from our engineering team where now that administrator is going to have to prompt, like go in and plug in root level access or admin level access.
So all of it, in my opinion, has been incredibly well crafted, very well done.
So one of the things I think people are confused about when they hear about agentic, they think it's a replacement thing.
The really thing that I'm finding as I'm setting up these systems is it turns your agents into a true force multiplier for the people.
I would also add on to that by saying what the VMware product engineering team has done here in terms of a very intentional way to execute this from a very safe lens, right, is making sure that this feature is completely compliant with any customer's kind of compliance regulations, right? Having it completely anchored in the data center, having it completely air-gapped if they need it to be, and self-efficient in terms of everything being located on-prem, but more importantly around the agent isn't going to be able to be this kind of toggled overachiever.
They're going to hunt that to the end, even if they have to blow up the data center or hack a frontier lab to accomplish it.
I mean, it's like you ask an agent today, outside of what we're doing, to go and execute a task.
if there's a piece of the task that they're not able to execute because they don't have line of sight or they don't have access, they're going to elevate permissions and they're going to figure out how to get that access to execute the end goal, which is the ask from the user.
And so these are all the areas in which our product engineering team has been very forward thinking around and ensuring does not happen.
So what they have really done in terms of nesting in existing, like, you know, carefully crafted skill sets for these agents, giving them the right level of constraints and guardrails so that even the IT end user can be very confident and they can trust this agent once they ask it a question and have like this, you know, long conversation around, you know, whatever it is they're trying to achieve, they'll always know and see what the agent is doing because there's that level of transparency and what it's executing, where it's looking for the information and doing the next steps.
But then also it'll be very confident in the follow through and know that if it needs to elevate permissions, well, that's a human level intervention that is very intentional from our engineering team where now that administrator is going to have to prompt, like go in and plug in root level access or admin level access.
So all of it, in my opinion, has been incredibly well crafted, very well done.
The rest of this transcript — segmented and speaker-labeled, so you land on the exact moment something was said
Search every transcript — by keyword, by phrase, or by meaning, across every show Radar indexes
Trends — what is surging across podcasts, measured against its own baseline
Alerts — when a name you follow appears in a newly indexed episode
No account is needed to search Radar.
