Case Study: Khan Academy accelerates AI debugging and observability with Langfuse

Khan Academy scales Khanmigo observability across 100 users and 11 teams with Langfuse

Khan Academy, a nonprofit educational platform with over 150 million users, faced a challenge integrating AI observability into its Go-based infrastructure. Most tracing tools required specific SDKs that were incompatible with their architecture. They turned to Langfuse's AI Engineering platform to build their flagship AI product, Khanmigo.

Langfuse provided a solution through its open API, which allowed Khan Academy's engineers to build a custom Go client. This integration gave over 100 users across multiple teams deep observability for rapid iteration and debugging. The results included increased development speed, hosting simplicity, and a unified platform across the organization, enabling Khan Academy to avoid the overhead of hosting its own tracing infrastructure. Langfuse was fundamental to their developers' understanding of AI implementations.

Khan Academy