If you are running an AI app in production, you want to give your users the latest and greatest features, while doing everything possible to ensure their experience is smooth and bug free. Decoupling deployments from AI configuration changes is one approach to reduce risk. LaunchDarkly’s new AI configs (now in Early Access) can help! With AI configs, you can change your model or prompt at runtime, without needing to deploy any new code. Decoupling configuration changes from deployment grants developers a ton of flexibility to roll out changes quickly and smoothly. An OpenAI API key is required for this tutorial, as well as a free LaunchDarkly account with AI configs enabled. The ExpressJS application generates letters of reference using the OpenAI model, which can be upgraded to a newer model using LaunchDarkly's AI SDK. The new model is deployed progressively to 25% of users, then 50%, and finally all users, while tracking metrics such as input tokens, output tokens, and satisfaction rate. Advanced targeting options are also available, allowing for more precise control over which users see which variation of the model. This tutorial provides a step-by-step guide on how to use LaunchDarkly AI configs to manage runtime configuration for an ExpressJS app.