Building Production AI Observability on Kubernetes: OpenTelemetry, Langfuse, Prometheus & Grafana From Zero to Scale
Building Production-Grade AI Observability on Kubernetes Most AI teams are solving the wrong problem. Everyone is benchmarking models. Very few teams are benchmarking infrastructure. Production failur
Responses
Markdown supported · ⌘↵ to post
Talha
Senior AI / Full-Stack Engineer | Building & running production AI systems
langfuse is quite good, the only issue is that they dont have gateway and tool versioning