
PromptQL is Hasura's AI data platform for querying, analyzing, and acting on enterprise business data.
PromptQL inherits Hasura's GraphQL Engine heritage, including fine-grained authorization alongside performance, scalability, and security that Hasura already operated at scale. The team describes the platform as batteries-included, with CI/CD and operational concerns abstracted away for developers.
A second advantage is the natural-language authoring model: business logic that previously required custom code can be expressed and shipped as PromptQL programs generated on the fly. This lowers the barrier to building data-rich AI applications on top of existing enterprise data.
PromptQL was launched by Hasura, the company behind the open-source GraphQL Engine, and is led by co-founder Tanmai Gopal. It is Hasura's first new data-access product after GraphQL and entered public beta with an announcement dated 02 June 2025.
Two days later, on 04 June 2025, PromptQL announced a partnership with UC Berkeley to develop a new data-agent benchmark for measuring the reliability of enterprise AI agents. The early collaboration signals a research focus on agentic data-access evaluation from the product's beta phase.
PromptQL is accessed through the Hasura console at console.hasura.io with a get-started-for-free entry point, reflecting a self-serve developer-console model. During its public beta the product is available to try without an upfront paid commitment.
This freemium console approach mirrors Hasura's broader go-to-market for developer infrastructure, where teams adopt the free tier and later move to paid enterprise plans. The strategy prioritizes bottom-up developer adoption during the beta period.
PromptQL is Hasura's AI data-access platform, which the company describes as the spiritual successor to GraphQL for the age of AI. It is Hasura's first new data-access product since GraphQL and is currently in public beta.
The PromptQL AI platform generates PromptQL programs on the fly, subsuming CRUD use cases while also letting complex business logic be authored and shipped in natural language. It is built so GenAI systems can interact with enterprise data without hand-written integration code.