We wrote up how our sandbox platform serving daily traffic on Perplexity handles millions of executi...

Aravind Srinivas on X: "We wrote up how our sandbox platform serving daily traffic on Perplexity handles millions of executions a day with a low error rate while still using multiple machines to lower serving costs." / X

[](https://x.com/)

Aravind Srinivas on X: "We wrote up how our sandbox platform serving daily traffic on Perplexity handles millions of executions a day with a low error rate while still using multiple machines to lower serving costs."

•  Aravind Srinivas @AravSrinivas We wrote up how our sandbox platform serving daily traffic on Perplexity handles millions of executions a day with a low error rate while still using multiple machines to lower serving costs. Perplexity @perplexity_ai 15h Our sandbox, SPACE, handles millions of executions a day with an error rate below 0.0002%. We’re sharing how it uses cryptographic identity to keep requests tied to the right execution as sandboxes pause, resume, and move. Read more: perplexity.ai/hub/blog/how-s… 4:57 PM · Oct 14, 2026·29K Views 34 4 197 43
•  Digital Foundation @digital_founder 14h The machines have formed a cost center support group [](https://x.com/compose/post?in_reply_to=2110416075888492680) 1 94
•  FutureAlpha @FutureAlpha12 14h An error rate below 0.0002% at millions of runs a day works out to a handful of failures. Most teams can't say that about a hello world script. [](https://x.com/compose/post?in_reply_to=2110415450291229066) 1 16
•  Winston B. @DoDataThings 13h Agent loops spin up a fresh sandbox on nearly every step, so p90 creation dropping from 447ms to 89ms hits the user's wait directly. Great work. [](https://x.com/compose/post?in_reply_to=2110425626893312228) 1 36

See all the replies

Continue to X

Log in or sign up for X

See what’s happening and join the conversation

Continue with phoneContinue with Apple

Continue with Google

通过 Google 继续操作 通过 Google 继续操作。在新标签页中打开

or

Log in with username or email