Human in the Loop: Two Ex-Microsoft Builders on AI
Sep 9, 2026 · 34 min · 10 segments
Chris has spent months trying to get his agents to talk to each other, and this episode is the honest tour of three attempts: what broke, what held, and what he actually runs now. He frames it against…
yeah you know i think we've been talking a lot right about like ai in the enterprise and so i i'm excited to to take a lot of what we're talking about and how can that scale into the enterprise right whether that's right building agents in a more enterprise platform, whether it's taking skills that we might define in cloud code or whatever harnesses and kind of publish them out to an organizational repo.
But then, of course, I think the topic that I've had lots of conversation with customers over the last year, year and a half, and that's gaining more traction, one around agents, but then to like, how do agents talk to each other?
And there's definitely situations where agents could communicate, work together with obviously oversight.
yeah man and i think towards the end of this right i want to talk about how you and i can improve our agents coming together i think ultimately like so a couple episodes ago we had renee on um for his for his second conversation and it was great right he talked about how his journey of tearing a system down rebuilding up with hermes but we were chatting with him afterwards and he was talking about this concept of getting his hermes agents to talk to claude code and that was just a light bulb moment for me i've taken and run with it but We'll talk about this more at the end if we have time, but like I would love for our agents to to meet.
I don't know if it's almost like taking your kids to a playground and like, hey, you want to bring your kids and see if they get along.
So we'll because I know you had an incident what last week where your Hermes agents.
I was trying to figure out how to run a second Hermes profile, a very scoped one for a very specific thing.
And your agent basically wrote a how-to guide and sent it over, which then again, Kyra was like, oops, yep, you're right.
yeah you know i think we've been talking a lot right about like ai in the enterprise and so i i'm excited to to take a lot of what we're talking about and how can that scale into the enterprise right whether that's right building agents in a more enterprise platform, whether it's taking skills that we might define in cloud code or whatever harnesses and kind of publish them out to an organizational repo.
But then, of course, I think the topic that I've had lots of conversation with customers over the last year, year and a half, and that's gaining more traction, one around agents, but then to like, how do agents talk to each other?
And there's definitely situations where agents could communicate, work together with obviously oversight.
yeah man and i think towards the end of this right i want to talk about how you and i can improve our agents coming together i think ultimately like so a couple episodes ago we had renee on um for his for his second conversation and it was great right he talked about how his journey of tearing a system down rebuilding up with hermes but we were chatting with him afterwards and he was talking about this concept of getting his hermes agents to talk to claude code and that was just a light bulb moment for me i've taken and run with it but We'll talk about this more at the end if we have time, but like I would love for our agents to to meet.
I don't know if it's almost like taking your kids to a playground and like, hey, you want to bring your kids and see if they get along.
So we'll because I know you had an incident what last week where your Hermes agents.
I was trying to figure out how to run a second Hermes profile, a very scoped one for a very specific thing.
And your agent basically wrote a how-to guide and sent it over, which then again, Kyra was like, oops, yep, you're right.
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.
