Postgres Summit US2026

PostgreSQL is usually treated as a passive datastore in AI systems while orchestration logic and workflow control live elsewhere in the stack. This talk presents an architecture where PostgreSQL becomes the control plane that governs AI workflows.

Large language models are inherently probabilistic, which makes them difficult to operate reliably in production environments. By placing a deterministic system around the model and using PostgreSQL to manage workflow state, validation, evaluation, and orchestration, we can constrain this stochastic core and make the overall system far more predictable and observable.

This talk introduces a practical design framework for building AI workflows around PostgreSQL. We will examine schema patterns for storing workflow state, validation and quality gates that prevent incorrect outputs from propagating through a pipeline, and evaluation loops that measure and improve model behavior over time.

Using a working implementation and practical examples, the talk demonstrates how PostgreSQL can orchestrate Retrieval Augmented Generation pipelines and agent workflows while enforcing deterministic control over probabilistic model behavior.

The talk will also include a short demonstration showing how an AI slide generation system can be orchestrated with PostgreSQL acting as the control layer. The same architecture can be applied to more complex enterprise AI workflows that require reliability, validation, and observability.