infrastructure for the agentic era. Orchestrate agents. Operate their environment. Automate workloads. Deploy what they build.
San Francisco, CA
- Tweets 367
- Following 2
- Followers 5,169
- Likes 253
Discord: rivet.dev/discord
92
GitHub: github.com/rivet-dev/rivet
1
102
Until now the only credential was the admin token, so actors sat behind your own API.
With scoped tokens, your backend signs the user in, asks Rivet for a token scoped to the actor, and the client talks to the actor directly.
1
209
Introducing JWT for Rivet Actors
Your frontend can now talk directly to actors — no proxying through your API
🔌 Supports Better Auth, Clerk, WorkOS, more
🎯 Granular access to actors & actions
🔄 Simple token refreshing
✨ Just 𝚖𝚢𝙰𝚌𝚝𝚘𝚛.𝚒𝚜𝚜𝚞𝚎𝚃𝚘𝚔𝚎𝚗(…)
2
4
33
4,215
Rivet retweeted
Running the harness outside of the sandbox is dirt cheap!
For example, Pi on Rivet Actors takes ~0.8 MB of RAM per active session
+ starts in tens of milliseconds
+ hibernates when idle
~~~
How?
When running an agent in a sandbox, you're paying for the overhead of an entire Node.js process for each session
That's usually hundreds of MB for every single agent session, plus multiple seconds to actually start up
When you run the harness within your own backend, you have a single process that can run hundres of agents per process
That means you've already paid for the fixed overhead of running Node.js (the hundreds of MB)
And each new agent session is just a few JS objects in memory
~~~
If you look at the graph, there's a baseline of 223 MB for the base Node.js process + imported libraries, but each new session takes only 0.82 MB
Granted, this depends heavily on how much of the session you keep in memory. Real-world results are going to be higher, but not by much since text doesn't take much memory.
Benchmark: github.com/rivet-dev/actors/…
~~~
In terms of security, the agent doesn't have access to the host it's running on
Instead, it's given tools that can talk to an external sandbox on demand
Read more about how to do this here: rivet.dev/blog/2026-07-27-ru…
how would it work if you want to run say 1000s of parallel pi sessions for muliple users? each of these is 200mb, no? how can you do that on your backend?
18
7
106
12,545
Discord: rivet.dev/discord
95
Read more: rivet.dev/changelog/2026-09-…
1
2
146
GitHub: github.com/rivet-dev/actors
1
3
198
Introducing OpenTelemetry for Rivet Actors & Workflows
🕰️ Trace actors that stay alive for days
🔗 Client → actor → actor in one trace
🗃️ Profile every SQLite query
🎚️ Per-action trace sampling
🦀 Rust, JS, and Effect support
2
4
51
5,678
Rivet retweeted
Secure Exec SDK got a huge refresh
(secure Node.js execution w/o a sandbox)
💅 Simpler API, host fns, TypeScript checking
⌛ 𝚌𝚛𝚎𝚊𝚝𝚎𝚅𝚖() for long-running processes & fs
📦 Dynamically install NPM pkgs
$ 𝚗𝚙𝚖 𝚒𝚗𝚜𝚝𝚊𝚕𝚕 𝚜𝚎𝚌𝚞𝚛𝚎-𝚎𝚡𝚎𝚌
5
2
49
2,260
BYOC is multi-region.
Each region gets its own VPC, cluster, operator, and control plane. Data stays in the region that produced it: pin EU users to an EU region for GDPR, or keep regulated workloads in-country.
Try it free for 14 days: create a BYOC project at dashboard.rivet.dev and follow the quickstart.
5
2,099
No inbound rules. Your admin token stays in your cloud secret manager. Rivet Cloud never connects in.
A Rivet operator runs in your Kubernetes cluster. It pulls commands from Rivet Cloud, applies updates to the control plane inside your cluster, and reports status back.
1
4
505
Introducing Rivet BYOC ☁️
for AWS and Google
Your data never leaves your security perimeter.
Your entire agent infrastructure is in your VPC.
Manage from the Rivet dashboard.
rivet.dev/changelog/2026-09-…
4
3
62
33,292
Introducing Rivet MCP
Connect your favorite agent to build, inspect, and manage your agentic infrastructure.
2
5
46
44,291
Your agent probably doesnt need a firecracker VM.
agentOS is 92x faster, 254x cheaper, and 42x smaller.
1
11
1,695
Discord: rivet.dev/discord
3
321