AI systems that can plan, use tools, and carry out multi-step tasks.
Trend this week
About the same13%of conversation
5 of 5 published weeks met the cross-podcast threshold.
12quoted moments6episodes4podcasts discussing this topic
A closer read
What changed
Its share fell from 15.9% to 13.4%, with evidence across 9 podcasts.
Where it showed up
It appeared in episodes including “BONUS: Snyk Found Malware Inside AI Agent Skills” and “Building an Autonomous Delivery Experience with DoorDash Co-Founders Andy Fang and Stanley Tang”.
How to read it
6 cited episodes from 4 podcasts are shown below. The topic appeared across 9 podcasts in the full analysis.
“When I'm kind of vibe coding using my agents, I'm already struggling, right? So like what changes are they making to my application? So these are the kind of new challenges from the developer point of view is, you know, like how does the security fit in with this new kind of d...”
“What could my agent know about what I want to eat or what I want to, um, buy from a grocery perspective? Like, help me understand, like, how you think about richer context or how to be smarter there.”
“Like it could be something as simple as, I don't know, like, like like dirt covering one of your camera sensors. Okay. Like, how does it— how robust is your autonomy stack able to, able to handle that? Like, there's some, there's some leaves on the ground, uh, but it only cove...”
“the easier we can make it for people to kind of interface with apps or with agents like they would with a person, I think it's going to reduce the friction in terms of compelling them to place an order”
“How do you think about building it with like working with a harness, right? So open code, Codex, you have your own pool CLI tool, basically getting people to use it, the co-design of model harness. Training it in?”
“we see incredible ability for complex harnesses that built lots of prompts around and extra data sources and other tools to really push capabilities of models forward.”
“5 or 6 tools per data source, and then there's the generic tools. And those are all read-only. And there are a few write tools that involve actually solving problems, like performing a code review, posting a comment somewhere, submitting a pull request.”
“I'm building a house and all that, and I needed something to keep track of all of my commitments and the argument I currently have with the contractor for the house. And I really wanted an agent that would kind of be like OpenClaw, but I didn't like OpenClaw at the time. I thi...”
going through the process of designing a complete microbusiness is going to have incredible learning effects for how you use AI in your normal context as well
Like, how do you even find the pin, the location of a customer's home? Uh, which again sounds kind of silly, but when you punch in someone's address on Google Maps, like the GPS pin, it's like, especially if you're going to apartment complex, it's never kind of, I mean, I mean...
Which I'll say, you know, you can see somewhat works looking at Anthropic. OpenAI has a lot of vision, multimodality. Anthropic kind of just did it, right? Fable's a big step up in image processing, but like, they're not known as the multimodal company, right? They're the lang...
Building an Autonomous Delivery Experience with DoorDash Co-Founders Andy Fang and Stanley Tang
Where, I mean, we, we also launched a product called Tasks a couple months ago where we're, we're having people in the Dasher fleet help basically collect, uh, data points to help train some of these world models. And I think we're so early there, and I think there's so many d...
Planet X. Uh, Palantir. Uh, uh, Alex Karp, when he's arguing that companies shouldn't use hosted frontier models that aren't open because they're handing off their alpha and their, their knowledge and so on and so forth. So it seems to me that GPT-3 is playing to exactly Karp'...
Its share fell from 7.6% last week to 3.8% this week.
Share of this week’s podcast conversation
Recent weeks6.5%→This week3.8%
Came up less often than in recent weeks.
Podcast moments5 cited momentsViewClose
The AI Native Dev
BONUS: Snyk Found Malware Inside AI Agent Skills
What can a developer actually do to make sure that any skills that they download, you know, is first of all safe? And how can an organization better adopt such technology with that governance in mind?
They did privilege escalation, moved laterally inside of OpenAI's own testing environment, figured out that Hugging Face happened to be hosting an actual solution to the evaluation, and instead of solving the test, they just chained together these um, I guess these stolen cred...
I thought it was interesting. So in the Laguna medium and extra small, you even quoted number of GPU hours for how many days and whatever for different size. And I'm like, oh, you can also work backwards to how much that costs, right? What GPUs, how many hours.
I think that's part of it. And then when it gets, uh, for everyone so popular— whereas an American capitalist company will undercut themselves with a loss leader until they got you hooked, and then they raise the price.
The tools drive another set of constraints and determinism. So I like that you're sort of building that. And I love also, like, as we had conversations about this, it also sounds like you didn't set out to do this, but in practice, this also saves you substantial cost, because...
