The alpha, twice a week
That was 3 min.
Get the best new moments in your inbox every Wednesday and Saturday. It's free.
Check your inbox
We sent a link to your address. Click it to finish signing up. Not there in a minute? Look in spam.
or
Latent Space · 3 min · Oct 27, 2026
Parallel agents need shared context and asynchronous work
Angela Jiang recommends a lead agent with parallelizable work and asynchronous sub-agents that do not block one another. She expects models to take on more architecture decisions over time.
Head of Product, Anthropic
Led product for Anthropic’s Claude Platform after leading API at OpenAI and Connect at Stripe.

Shawn Wang (swyx), Host · Co-founder, AI Engineer
Wrote “The Rise of the AI Engineer” and founded Smol AI, maker of AI News.
About this segment
Angela Jiang recommends a lead agent with parallelizable work and asynchronous sub-agents that do not block one another. She expects models to take on more architecture decisions over time.
From “How Anthropic ships the #1 LLM in the World — with Katelyn Lesse and Angela Jiang”, Oct 27, 2026. Watch the full episode
Thanks. We'll look at and fix what's off.
“That is like the only two bits I would say you should care about in your architecture.”
Angela Jiangat 1:45:45 · Latent Space1:45:45Takeaways
- Give one lead agent work that can be split in parallel.
- Keep sub-agents asynchronous and non-blocking.
- Shared context helps parallel workers make progress.
In this episode
Full episode- 56:02Claude helped cut Messages API CPU use by up to 50%3:19
- 1:03:18Separating the agent from its sandbox limits security risksPro pick 4:56
- 9:41Anthropic makes reasoning harder to distill without breaking API use3:27
- 19:08Model launches now include policy and authorization work2:54
- 31:56Early-access model testing depends on customers with robust evals3:41
- 1:20:59Humans own system design for Claude-written code3:54
1:43:26Parallel agents need shared context and asynchronous work3:14
Transcript
From the transcript Latent Space publishes with the episode.
- 1:43:29
That, that was my soapbox. Uh- [laughs]... uh, you know, closing questions. You've been very generous with your time. Any parting thoughts or call to action? You start. Build something cool on Opus 5.5. What's something- Yeah... what's something people should try to push the bounds with new Opus 5.5? You guys have been using it for the longest. Yeah. I mean, it's really cost efficient. Um, it's less verbose.
- 1:43:49
All the things we mentioned. The vibes are really, really high. It's been fun. Uh, yesterday was a really fun day for us 'cause like, uh, finally again I got like a s- a whole slew of texts from friends- [laughs]... and, and people who were just like, "Amazing model. Thank you." Like one of my old bosses, um, messaged me, and he was like, "Opus 5.5 just saved our butts in an incident," right? And it's like, okay, great, like we're, we're back, is really the energy that we feel.
- 1:44:11
And so I think things that like maybe you wanted to experiment with something and like cost was something that was front of mind for you, right, and like token efficiency and these sorts of things, like that kind of might have held you back from like really like firing it off and like letting a model run. I think that's the... Those are the sorts of problems you should return to now with Opus 5.5 'cause I think you'll be able to get just like a lot further than you maybe otherwise would've been able to with, um, models that were less cost effective.
- 1:44:38
I think you should try something with like a multi-agent setup. I think you will find it do very good. Any, any other hints, by the way? I, I, I mean, multi-agent's very broad. Uh, nope. Okay. I think it'd be very curious when we hear what people build. Um, but yes, that it's, uh, it should be extra good, so. Yeah, including, let's say, uh, roles that are not strictly defined, but they assign among themselves.
- 1:44:58
Mm-hmm. Right? Which is, uh, to me that the new thing. Like- Yeah... you can assign roles, you could always assign roles. Mm-hmm. Uh, you can give tokens jobs. Um- Indeed... but, uh, this is like they're giving themselves jobs. Yes. Yes. I guess, um, say more on multi-agent, right? Like, should we use sub-agents? What do you... How should people think about multi-agents? This is pretty good in a sense of like, basically, generally, you wanna have one lead, and you wanna like split up like a bunch of different stuff.
- 1:45:25
And as long as you can have a task that you think could be generally parallelizable, and they share uniform context across it, I think you will get a lot of gains. Now, of course, it depends on the problem that you're pushing at, but like increasingly, this is a very superior way for us to handle a lot of these things, and especially if every sub-agent is, or whatever you wanna call these, like sub-leads, they are asynchronous, um, and they're not blocking of each other.
- 1:45:47
That is like the only two bits I would say you should care about in your architecture. Other than that, like let it rip and like see what Claude designs for you. And increasingly, I think a lot of this architecture stuff will fall into this direction. Like Claude will get smarter and smarter. It's going to learn how to figure ways to like split itself, clone itself, adjust itself to the architectures that make sense. You just need to prevent it...
- 1:46:06
You just need to make sure it can like parallel process, so it needs to share state, and it needs to like not block its sub-agents. And as long as you contain those two principles, honestly, I think people should experiment and probably try some really crazy stuff. And, uh, I wouldn't be surprised to see some like really interesting things on top of that, that maybe they felt like there was some intractable problem that was really annoying and they keep pushing for, that something like this could actually probably just like crush in maybe one shot or something.
- 1:46:29
Excellent. Yeah. All right, everyone go build. Cool. [laughs] Cool. [outro music]