July 2024 Summaries

HMH, a 100+ year old textbook publishing company, partnered with Hasura to improve their digital classroom experience. They faced the challenge of managing limited classroom bandwidth for 30+ users, which led them to look for innovative ways to structure their tech stack and APIs. Hasura's low-code, YAML metadata-based configuration allowed HMH to compress development and feedback cycles immensely, reducing time-to-market by over 3x and API development and testing time by 5x. With Hasura, HMH was able to streamline internal APIs with Snowflake, power their interactive classroom application ClassCraft, and scale to meet future needs, positioning them for continued innovation in the ed-tech space.

GraphQL is a query language for APIs that allows developers to ask for exactly what they need in a single request, giving them the ability to iterate independently of backend developers and fulfill their dreams of harmonic convergence. However, it has reached a plateau where the weather is mild and the terrain is gentle, and pioneers must plan a difficult descent and journey beyond. Despite rumors of its demise, GraphQL is far from extinct, but environmental conditions threaten its path to flourishing, including security concerns, performance issues, and complexity. To address these challenges, some argue that GraphQL should be rethought as a development tool, rather than a query language, and implemented with a compiler for the underlying data platform's native query interface. This new approach eliminates the N+1 problem, paves the way for powerful strategies for implementing authorization, and generates efficient and type-safe code. By embracing this new perspective, GraphQL can flourish and overcome its perceived problems, allowing it to move beyond its plateau and continue to evolve as a technology.

A leading European sports betting company reduced its API delivery time with Hasura, a cloud-based API platform, allowing it to launch new features faster and focus on more impactful work. The company's data team collects, transforms, and aggregates customer-oriented data into a centralized PostgreSQL database, making it available to the broader technology group via APIs. With Hasura, the company can model its data using SQL, eliminate most performance tuning and debugging issues, and set up fine-grained permissions for secure access control. Composability enabled by Hasura also freed up time on the data team, allowing them to enable self-serve data access without involving them in every request. This has resulted in a dramatic acceleration of new feature launches and improved overall customer experience.

The Hasura Turso connector is a tool that allows developers to connect their SQLite databases to GraphQL APIs, providing a flexible and efficient way to manage data. With the connector, developers can run various queries and mutations on their SQLite databases, including simple queries, queries with predicates, ordering, and pagination, as well as querying relationships and performing mutations such as insert, update, and delete operations. The Turso connector also supports additional operations like delete many, insert many, and update many, and will soon include capabilities for aggregations and grouping. It can be used to try out the connection with Hasura DDN and is documented in a GitHub repository.

The Hasura Qdrant connector is a powerful tool that enables developers to perform semantic searches on their data using GraphQL, specifically vector and recommendation search, which are increasingly important in Retrieval-Augmented Generation (RAG) workflows. The connector allows for simple queries, queries with predicates, queries with vector search, queries with vector search and predicate filtering, queries with recommendation search, insert mutations, update mutations, delete mutations, and the ability to create relationships between Qdrant and other data-sources such as PostgreSQL. This powerful feature enables developers to perform a vector or recommendation search and then join the resulting data back to their PostgreSQL database, enhancing the accuracy and relevance of information retrieved by AI models.

The Hasura DuckDB connector provides a GraphQL query layer for DuckDB databases, allowing users to easily expose their data through GraphQL queries. It supports various types of queries, including simple ones, those with predicates, ordering and pagination, and complex queries that utilize relationships. The connector also allows filtering across relationships. With this connector, developers can leverage the power of GraphQL for their data stored in DuckDB databases, whether it's a local instance or one hosted on MotherDuck.

A Fortune 100 U.S. bank has successfully implemented Hasura to address critical challenges in its data governance and reporting processes, improving control of its data while reducing costs and ensuring accuracy. The bank's existing architecture suffered from expensive manual processes, error-prone data, and regulatory risks, but Hasura delivered several benefits including low-code API development, domain-driven metadata, granular access control, private deployment, next-gen observability, and standards compliance. By implementing Hasura, the bank expects to decrease manual work, minimize reporting errors, enforce proper data usage, and position itself for future growth and regulatory compliance in an increasingly data-driven financial landscape.