Data APIs and enterprise AI
Hasura
Hasura builds software that connects applications and AI to organizational data. Hasura GraphQL Engine and Data Delivery Network (DDN) generate GraphQL APIs with fine-grained authorization, helping developers build applications over existing databases and services. Its expansion into PromptQL in 2025 brought natural-language data queries, business logic, and workflow automation into the product family, while retaining maintenance of HGE and DDN. PromptQL gives teams shared threads and a shared wiki for business context, with permissions enforced at the data layer.
Founded by Tanmai Gopal and Rajoshi Ghosh, Hasura grew from tooling developed for a planned food-delivery business into a company focused on simplifying application development. Gopal is CEO of PromptQL. Hasura’s enduring engineering approach uses declarative metadata to describe data mappings, relationships, and authorization rules. Its query engine compiles GraphQL requests and reduces downstream data fetches, while integrating permission checks into data access. PromptQL extends that foundation to AI working with shared organizational knowledge.
In 2022, Hasura raised a $100 million Series C led by Greenoaks at a $1 billion valuation, bringing cumulative funding to more than $136 million.
2 talks
Newest first
Building efficient hybrid context query for LLM grounding
AI Engineer Summit 2023 · 11:35
2 speakers at AIE
Affiliations reflect their AIE appearances, not necessarily current employment.
Start here
- Building efficient hybrid context query for LLM grounding
Start here to learn how a Hasura GraphQL API combines semantic retrieval with structured filters, using Weaviate and Postgres in a product-search demonstration.
Simrat HanspalAI Engineer Summit 2023
- Hasura Launch: Realtime Data Connectivity for AI
Use the customer-email example to understand how Pacha DDN proposes connecting LLMs to business logic and live information across multiple sources.
Tanmai GopalAI Engineer World's Fair 2024
Messages from the stage
Permissions around generated queries
Simrat Hanspal's demonstration pairs OpenAI-powered query generation with restrictions on application permissions to prevent unintended operations.
Schema-aware retrieval planning
Tanmai Gopal outlines a unified SQL-like interface across structured data, unstructured data and APIs, with schema-aware rules and generated Python code for autonomous retrieval planning.
Affiliations reflect each recorded session, not necessarily current employment.