We're building the next generation of PostgreSQL infrastructure.Explore our roadmap

PLATFORM FEATURE

Embeddings belong beside your application data.

Planned pgvector support for embedding storage and similarity search within a familiar relational data model.

Private alpha target

At a glance

PostgreSQL-native vector extension
SQL alongside vector retrieval
Tenant-aware security controls
01 / INSIGHT

Designed around developer workflows

UnDataly is building this capability with explicit security and operational gates. The details below describe intended developer scenarios, not a currently available public API.

  • Store semantic embeddings with records.
  • Prototype retrieval-augmented experiences.
  • Combine structured filters with similarity search.
02 / INSIGHT

Availability and release readiness

Status: Private alpha target. Readiness depends on verification of isolation, authorization, recoverability, reliability and cost controls. No availability date is guaranteed.

Frequently asked questions

Can I use UnDataly Vector in production today?

No. UnDataly is in engineering development and is not accepting production customer workloads.

How can I learn when it becomes available?

Join the early-access interest list for announcements about the rollout.