Aug 19, 2026 · 1 hr 21 min · 15 segments
Opus 5 has started taking actions nobody asked for, GLM-5.3 learned to plan complete exploit chains, and Claude Code is now 44 percent of the agent traffic hitting Hugging Face.
okay uh let's see here uh where do we want to start aws news for the week and a half um The first thing I have on my list here is that you can now...
We talked about this, I don't know, maybe a month ago? I don't know.
In the past.
It's kind of like an episode idea.
I have no idea.
Remember we talked about the idea of using...
Oh,
sure.
Yeah.
Yeah.
It now also supports and I had to look up what the heck this is.
Bedrock Mantle.
I was like, is this a new service? Like what is Bedrock Mantle? yeah i've never heard of that yeah you know it's funny that you say that because i think we may have talked about this once very briefly and i think we both went yeah that's cool and moved on and then and then uh i was like oh man what is mantle so i went and i looked and essentially mantle is um native drop-in compatibility with open ai's apis that's all it is
okay yeah so that's like Specifically their agent tools or whatever services they're offering on their system.
Okay,
got it.
You can talk to them directly, same API calls and everything that you're used to.
So Mantle now also supports this idea of cost allocation through Bedrock.
And I can tell you...
That has been very helpful for me, not the mantle piece, but just the cost allocations.
So I've gone through and tagged, tag isn't the right word.
I have a bunch of Lambda functions, right? Those Lambda functions all have individual principles, roles.
And then what I've done is in the cost allocation, just like any other cost allocation thing, you've got a feature.
okay uh let's see here uh where do we want to start aws news for the week and a half um The first thing I have on my list here is that you can now...
We talked about this, I don't know, maybe a month ago? I don't know.
In the past.
It's kind of like an episode idea.
I have no idea.
Remember we talked about the idea of using...
Oh,
sure.
Yeah.
Yeah.
It now also supports and I had to look up what the heck this is.
Bedrock Mantle.
I was like, is this a new service? Like what is Bedrock Mantle? yeah i've never heard of that yeah you know it's funny that you say that because i think we may have talked about this once very briefly and i think we both went yeah that's cool and moved on and then and then uh i was like oh man what is mantle so i went and i looked and essentially mantle is um native drop-in compatibility with open ai's apis that's all it is
okay yeah so that's like Specifically their agent tools or whatever services they're offering on their system.
Okay,
got it.
You can talk to them directly, same API calls and everything that you're used to.
So Mantle now also supports this idea of cost allocation through Bedrock.
And I can tell you...
That has been very helpful for me, not the mantle piece, but just the cost allocations.
So I've gone through and tagged, tag isn't the right word.
I have a bunch of Lambda functions, right? Those Lambda functions all have individual principles, roles.
And then what I've done is in the cost allocation, just like any other cost allocation thing, you've got a feature.
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.