Dakera Deployment
Deployment configuration for self-hosting Dakera, an AI agent memory server that combines vector search, hybrid BM25 and HNSW retrieval, knowledge graphs and session management in a single Rust binary, covering Docker Compose, Kubernetes and Helm.
This repository packages Dakera, a self-hosted memory platform for AI agents, for deployment with Docker Compose, Kubernetes or Helm, including a high-availability cluster setup and monitoring. Dakera itself is a single Rust binary providing vector search, hybrid BM25 and HNSW retrieval, knowledge graphs, session management and built-in embeddings with no external dependencies, so operator data stays on the operator's own infrastructure.
Described use cases include shared memory across multi-agent pipelines, retrieval-augmented generation with long-term context, persistent memory of past decisions for code-review or DevOps agents, per-user session history for support bots, and long-term preference storage for personal assistants; all use cases deploy the same way, with the deployment profile choosing the storage backend and scale.
A local profile runs entirely in memory for quick testing with a single Docker Compose command; a Development profile adds MinIO-backed persistence, and a Default profile is meant for production. A devcontainer configuration supports IDE-integrated development in VS Code or Cursor.