
E2B
1.6K posts

E2B
@e2b
Machines for AI agents | Hiring! 👉 https://t.co/WTLBOEvNRx


Real GRC work ≠ simple Q&A. It means reasoning across frameworks, controls, and data with real complexity... more than any context window can hold cleanly. We gave the Vanta Agent a sandbox so it can go from describing the work to actually doing it: real reports, real exports, real assets. Lavanya Kannan broke down everything we learned, including what worked (and what didn't), in our latest installment of Trustcraft. Link in comments 👇




Filesystem-only snapshots are now available on E2B. Filesystem-only snapshots skip capturing memory and process state, making them quicker to create and restore while using less storage than full snapshots. Use them when you only care about the data on disk and don’t need running processes restored. For example, to persist a workspace between sessions or save build and scratch state without restoring running processes. Docs in the thread.





From the beginning, E2B has consistently ranked near the top of the pack on performance, while also providing the most complete form of workload isolation (full microVMs), most complete lifecycle capabilities (pause/resume, snapshot, etc), and a diverse set of deployment options (BYOC). We've never needed to benchmaxx because the years of R&D we've put into fast and secure machines tend to take care of that on their own. Nonetheless, we participated in ComputeSDK’s 100k Burst Concurrency Benchmark and crushed it as the fastest microVM provider with a 100% success rate across 109k sandboxes. We just keep doing what we do as the first, fastest, and most reliable Agent VM provider there is. Extremely proud of the team for being able to achieve this. Benchmark is linked in the comments.

Our own Matt Brockman is hosting a workshop on top 5 sandbox failure modes and how to address them at AI Engineer! Catch his workshop to become an expert in managing sandboxes in just 1 hour. You'll trigger & patch various failure modes live: burst traffic, callback loss, state leakage, runaway execution, and brittle templates. Join us at @aiDotEngineer on Monday, June 29 at 11:05am at Moscone West.





New banger integration alert: you can now use E2B sandboxes as the backend for your @LangChain Deep Agents. Get started by installing the dedicated package with pip install langchain-e2b, then follow the LangChain Deep Agents docs for step-by-step integration.






