Historique et présentation rapide de la problèmatique de single source of truth entre backend et frontend. Rappel des solutions existantes et des limitations (codegen, openapi, graphql, trpc...). Présentation de AWS smithy.io et de la modélisation de service. Le codegen client et server avec la CLI smithy. Mon implémentation pour intégrer le server sdk dans Nest.js.

Software Architects choose tools and patterns with an eye toward cost, time, shelf-life, and standard features. Data API builder is not source generation — it's Microsoft's free Data API engine that exposes secure REST & GraphQL endpoints over SQL Server, Postgres, MySQL, and Azure Cosmos DB, whether in Azure, on-premises, or even in AWS or GCP. In this session, we'll discuss security & architecture options, and when to use GraphQL versus REST endpoints.

Jian Huang speaks on GraphQL+Graph databases usage and how we leverage Kafka, Elastic Search, Hackolade together.

Is GraphQL still the game-changer we thought it was in 2018? In this talk, we'll take a critical look at where GraphQL stands today. We'll explore how other technologies have evolved and why our current approaches might not be cutting it anymore. Are we solving round problems with square solutions? Are we overfitting data access to a GraphQL shape? Let's dive deep on the pitfalls of GraphQL in 2024, how to mitigate them, and get the best that GraphQL has to offer with future-proofed APIs.

GraphQL can be challenging, and not only for beginners. The reason for this is often that it’s so tempting to do the wrong thing. In this session, I will walk you through the various challenges of building production-ready applications with GraphQL. We will discuss the differences between a public GraphQL server and a private GraphQL server. I will show the most common mistakes people encounter with GraphQL and how to tackle them. In this talk, we will look at persisted operations, rate-limiting, introspection security, the right defaults for paging, errors, and many more aspects. This talk is suitable for you whether you are a GraphQL server or client implementor or just a user of GraphQL. After this session, you should have a solid understanding of how to put GraphQL into production in a secure way. Michael Staib is a member of the GraphQL technical steering committee, a Microsoft MVP, and the author of the Hot Chocolate project (https://github.com/ChilliCream/hotchocolate), a platform for building GraphQL servers and clients in .NET. This open-source project has been his main focus for the last couple of years. Apart from his work in the open-source community, Michael works as a consultant to help companies move to GraphQL. You can tune into the Hot Chocolate project on youtube: https://www.youtube.com/@chillicream

Data is power, but building APIs is tedious. Engineers create vital value by modelling domains and data, but waste time on repetitive plumbing tasks like CRUD, data pipelines, and cross data source joins. What if you could skip all that? With supergraph, you can. Query the supergraph with GraphQL, and get consistent features like joins, filtering, and aggregations across all data sources. Set permissions where they belong: at the model level, where they can be applied to absolutely any query. Live demo of how to build a supergraph that connects the GitHub API with a users database will be presented.