LaunchDarkly

@LaunchDarkly

Runtime control for the AI era. Helping engineering teams ship, govern, and control code and AI agents in production without waiting on a redeploy. 🚀🧑‍🚀

Oakland, CA

launchdarkly.com

Replies Reposts Media

LaunchDarkly
@LaunchDarkly
2h
The LaunchDarkly AI SDK has just landed 🚀 Available now for Python and JavaScript, working with the providers and frameworks you already run. Metrics recorded automatically, traces one package away, and full agent workflows from a single call.
00:00
2
LaunchDarkly
@LaunchDarkly
Sep 9
Somewhere in your production environment, an agent just did something nobody signed off on. Our hands-on workshop on September 17 is about getting control of that before it becomes a problem. Think model swaps without a redeploy and rollouts that catch issues on their own.
1
LaunchDarkly
@LaunchDarkly
Sep 8
Heading to HumanX Amsterdam! Plenty of people can demo AI, but we want to talk to the ones running it in production with real users on the other end. See you there! 👋
1
LaunchDarkly
@LaunchDarkly
Aug 28
A routine cleanup broke prod within minutes. One of our engineers pushed a small code change last summer that he was sure was safe. Before most users ever saw the bug, our own factory caught it and rolled the change back automatically.
4
LaunchDarkly
@LaunchDarkly
Aug 27
to do: stop finding out prod broke from a Slack message Latest Control Panel showed flags that heal themselves, an agent that cleans up stale flags and opens the PR for you, and experiments that survive a mid-run metric change.
You can't control what you can't see | LaunchDarkly
From launchdarkly.com