Derek Collison retweeted
What happens when a company's agents aren't all running in the same place? Or aren't using the same harness?
Durable Execution should not break the bank.
youtube.com/watch?v=TWb3G1CA…
1
16
1,115
Want to learn about @synadia Agent Fabric? Here is a great list of videos covering all you need to know.
youtube.com/playlist?list=PL…
Impressive. I do believe in a world where design, architecture and very importantly "taste" is still human driven, but AI does the rest.
@mitchellh and team are doing this very well. Great stuff.
Rex demo time: magical and automatic remote copy/paste, drag/drop, and port forwarding (e.g. open web servers running on remote machines in your local browser). Plus, showcasing some of the internal plumbing that powers this and how you can hook into it with Lua scripting.
Our goal with Superlogical and Rex is for everything to feel local. I've said before: the moment any UI shows a selector between "local" and "cloud", the product has failed. The lines should blur, and we should build the foundational technology that makes it happen performantly, securely, and beautifully.
Great place to start is this video to get a quick understanding of the main concepts:
Identity, Discovery, Connectivity, Tracing, and State/Context sharing.
youtube.com/watch?v=iNIHCn7W…
Most agents today have one caller - the person typing to them.
The next ones will take requests from pipelines, services, and other agents, and nobody will be watching.
Synadia Agent Fabric is built for that second kind of agent.
Every prompt it receives is signed, and every request lands on one record you can follow.
Public preview of Agent Fabric, all built on @NATS_io, now available
The @nats_io ecosystem continues to grow, which is great to see.
Moving a stream between @NATS_io nodes is easy when the cluster is healthy: add the new servers, copy the data, remove the old ones.
But you often move a stream because something is wrong. A node is failing, lagging, or about to be retired.
Getting those steps right on a degraded cluster took a complete redesign in NATS Server 2.15.
Before 2.15, if a node dropped out at the wrong moment, the system could keep the wrong replica and lose data. 2.15 fully closes that gap.
Now, if a server drops mid-change, the change waits. It never takes a shortcut with your data.
As AI agents become more specialized and increasingly distributed across physical locations, providers, and ecosystems, the need for secure identity, seamless connectivity, and the ability to share context and state becomes critical. @synadia believes this will be the foundational infrastructure required for the next era of agentic AI.
Soon, most of the prompts to your agents will come from software, not from people.
Synadia Agent Fabric, now in public preview, signs every prompt, gives each agent its own model key, and puts every call in a task on one timeline.
All running over @nats_io
How do AI agents coordinate across clouds without losing state? Andrew Connolly and @wallyqs talked NATS and AI workloads with @theCUBE at #KubeCon 🎥 youtube.com/watch?v=Yih7BxZd…
Huge congrats! So excited to be on this journey with you and the extended team.
If you see one on the road, there are @nats_io servers in there!
R2 is the 2027 @MotorTrend SUV of the Year! 🏆
Ed Loh, Head of Editorial for MotorTrend, said, “Rivian built vehicles engineered with a clear understanding of what drivers want: a balance of thoughtful design and useful capabilities.”
Read the full story at rivn.co/r2-motor-trend-2027.
A big congrats to all the @NATS_Io maintainers, contributors, advocates, end users, and community members who have shaped the NATS ecosystem over the years.
xitter.gq/cloudnativefdn/status/…
NATS has officially graduated from CNCF.
The open source messaging and streaming fabric now provides a proven communication layer for more than 2,000 organizations running distributed systems and agentic workloads.
Read the announcement:cncf.io/announcements/2026/1…
#KubeCon #CloudNativeCon
There's a new version of this post
Such an amazing journey! @nats_io just turned 16 as well.
A huge thanks to everyone in the ecosystem for your support.
It feels like we are just getting started. At @synadia we have never seen so much interest in NATS as the central nervous system to drive modern distributed systems, cloud to edge, and of course Agentic AI.
NATS has officially graduated from CNCF.
The open source messaging and streaming fabric now provides a proven communication layer for more than 2,000 organizations running distributed systems and agentic workloads.
Read the announcement:cncf.io/announcements/2026/1…
#KubeCon #CloudNativeCon
There's a new version of this post
Derek Collison retweeted
October in TigerLand is out!
- Official Rust client published!
- State Machine Fuzzer rewritten
- The Tale of TigerStyle x Object Storage
tigerbeetle.com/newsletters/…
Great outcome!
We just replaced Firestore subscriptions at @Lovable with a system built on NATS, cutting application payload by 97% compared with sending full results.
Over 2 million concurrent subscriptions at peak. An estimated 40 TB → 1.11 TB actually sent in 30 days.
What’s crazy is that @AtaMarzban94 built the core in 40 days, and we rolled it out without a single user-facing incident.
Using proprietary software that 80% fits used to be fine, now you can just build your own that fits you perfectly. It’s like why would someone buy somewhat-fitting mass-produced clothes if they could easily make their own bespoke?
Derek Collison retweeted
At @Lovable, we reduced realtime data transfer from an estimated 40 TB to 1.11 TB per month.
@AtaMarzban94 built the core in just 40 days.
Really cool engineering story.
xitter.gq/AtaMarzban94/status/21…
2.6 billion updates in 30 days. Over 2 million concurrent subscriptions at peak.
That’s @Lovable Realtime, the system keeping our browser tabs in sync.
I built the core in 40 days. We rolled it out to every user without a user-facing incident.
Here’s how it works.
.@Lovable keeps 2M+ live subscriptions fresh with @NATS_io—and the code that writes the data never has to trigger a refresh.
Every database change gets published to one specific subject in the hierarchy. Each subscription listens with wildcards, as broadly as it needs, and NATS routes each change to the subscriptions it affects.
For a Lovable dev, a live feature is a backend function written like an API endpoint, plus one hook call on the frontend.
They have ~40 of these today, and teams ship new ones independently!
Check out the full writeup from @AtaMarzban94 below👇️
Good read! It has been great working with @Lovable and making sure that @nats_io drives the value that they need in such a fast paced environment.





















