Supabase: five million databases a month as agents lift conversion

Supabase's product announcements at its Select event were grouped into Build, Operate and Scale, and the practical thread running through them is removing work that developers have done for years. Running Supabase locally without Docker removes another dependency from the path to a working database, and declarative schemas put the desired state in Git so an agent can read, change and fork it alongside the application. The hosts compare that shift to a Terraform moment for the database.

On the operating side, logs, checks, connections and context help an agent understand what it is changing; a model that can issue SQL still needs to know what the database contains and what happened when something failed. Supabase Compute, in private alpha, addresses the place to run processes like document chunking or embedding generation, reducing the number of separate services to configure and explain to an agent, while edge functions remain useful but do not cover every workload.

The Scale announcements — OrioleDB, Multigres and the Turso acquisition — reflect two different engineering problems: serving more small databases and supporting a customer's increasingly large database. The founders' observation that AI-driven usage improved activation and conversion sits alongside that breadth, and their explanation is that an agent can carry a user from signup to a working setup in one flow.