Tiger Data, the company behind TimescaleDB, has launched Agentic Postgres, a Postgres-based database designed for both AI agents and developers. It extends Postgres with fast forking, an MCP server, native BM25 and vector search, and includes a CLI for terminal access.
Agentic Postgres’ MCP server allows agents and developers to interact with the database through high-level prompts such as: “I want to create a personal assistant app. Please create a free service on Tiger. Then using Postgres best practices, describe the schema you would create.”
Tiger Data says they have distilled more than a decade of Postgres experience into a set of built-in master prompts covering schema design, query tuning, migrations, native full-text and semantic search, and more.
Full-text and semantic search are powered by two Postgres plugins: an improved version of the existing pgvectorscale, which delivers higher-throughput indexing, better recall, and lower latency at scale; and the new pg_textsearch, which implements BM25 for modern ranked keyword search and is optimized for hybrid AI workflows. Currently, pg_textsearch runs in-memory to ensure maximum speed, with disk-based operation support in development.
As Tiger Data’s CTO explains, at the foundation of Agentic Postgres lies Fluid Storage:
A new distributed storage system built for elasticity, iteration, and safety: a transactional distributed block store, lineage-aware storage proxies, and a user-space block device driver, with versioned copy-on-write and application-consistent snapshots coordinated with Postgres.
Thanks to Fluid Storage, Agentic Postgres delivers fast, zero-copy forks on real production data, which makes it possible to create instant environments, to experiment safely in sandboxed environments, and to run agentic loops directly on real data. For example, an agent can spin up its own isolated environment with a full copy of production data in just seconds, says Tiger Data, and test whether creating new indexes would improve performance.
Contrary to being merely a matter of efficiency, Tiger Data emphasizes that fluidity, i.e., true elasticity or the ability for a database to scale, fork, and contract instantly, is critical for agentic software. In a typical workflow, agents create, modify, and deploy code autonomously, run migrations, benchmark results, and tear everything down again. To be effective, this must be accomplished in seconds. According to Tiger Data, this level of performance cannot be met by existing database services like Amazon EBS, which introduce higher latency on replicating a database and impose limits when scaling down or resizing.
As we began exploring agents for both internal and customer use, it became clear that today’s cloud infrastructure needed to be rethought.
Level Up Coding founder and software engineer Nikki Siapno noted on X.com that AI is changing what developers require from databases:
They now need to handle time, meaning, and memory. All in one place. Traditional systems weren’t built for that. […] Agentic Postgres by Tiger Data brings time, meaning, and memory together.
She admits that you could “stitch together” add-ons or multiple specialized databases to satisfy the requirements of agent memory, time-series at scale, vector search and embeddings, semantic retrieval, and others, but that “only adds complexity”.
Other database services that specifically target agentic software are Firebolt, which is Postgres compatible and optimized for analytical workloads with high concurrency and sub-second query latency, as well as Weaviate and Qdrant, which support high-dimensional vector storage, similarity search, and metadata filtering.
Developers can sign up to try out Agentic Postgres using a free tier which gives access to forkable databases, hybrid search, memory APIs, and MCP integration with limited bandwidth and performance.
