LLM Observability Platform - Production Deployment Suite
A self-hosted infrastructure suite that deploys a multi-service observability pipeline for LLM applications, covering cost, token and semantic-cache metrics across Postgres, ClickHouse and Redis, with Kubernetes and Terraform options.
LLM Observability Platform is a self-hosted deployment suite that installs and orchestrates infrastructure for observing large-language-model applications, running from a local development machine up to multi-cloud Kubernetes clusters. It ties together a set of backing services, including Postgres, ClickHouse and Redis, behind automated schema migrations and scheduled backups.
It generates network policies and firewall rules that isolate the stateful data services from public networks by default, and provides Terraform configuration for AWS, GCP and Azure alongside Docker Compose and Kubernetes deployment paths. Infrastructure is managed through shell scripts and a command-line orchestrator that start, stop and inspect the containers, and an application layer of Node.js microservices sits on top of the infrastructure.
The project documents minimum and recommended hardware for local development versus production, and treats semantic caching and OpenTelemetry-based tracing as part of the observability data it collects.