October 2024 Summaries

Firebase App Hosting, a secure serverless platform for hosting server-rendered web applications, has expanded beyond its initial us-central1 availability to asia-east1 in Taiwan, with europe-west4 in the Netherlands planned shortly. Developers can select regions through the Firebase console or CLI, potentially reducing latency by locating services closer to users and isolating regional failures. The platform also now supports manually triggering deployments from GitHub branches or specific commits through the Firebase CLI, allowing integration with custom CI/CD workflows instead of relying solely on automatic GitHub-based rollouts. Other recent updates include CLI project initialization, rollout-status polling in the console, dynamically scaled Cloud Run settings, Next.js 15 support, and a fix for cookie header handling. Firebase plans to introduce an App Hosting emulator for local development and continue adding regions and features.

Firebase Crashlytics has introduced a public preview of enhanced Gemini AI Assistance that lets developers provide code and additional situational context when investigating mobile app issues. Previously, Gemini generated one-time issue insights using only Crashlytics data, but the updated interface supports follow-up input through separate fields for relevant code and other details, then regenerates a response using the original crash data, prior output, and supplied context. Google reports that this added context can produce more accurate diagnoses, targeted code recommendations, suggested fixes, and code-quality guidance, while emphasizing that generated code should be tested before production use. Developers can access the Code Context beta after configuring Gemini in Firebase and receiving the required permissions, using the “Get Started” option beneath an initial issue insight.

Firebase Data Connect, now renamed SQL Connect, uses GraphQL as the interface for a Cloud SQL-hosted PostgreSQL backend, generating client APIs, type-safe SDKs, authentication integration, and tools for generative AI workflows from application schemas and operations. GraphQL was selected for its concise, strongly typed, language-agnostic, customizable, widely supported, and mature design, as well as its ability to handle relational queries and filtering. Developers define database structures with GraphQL types and directives, while SQL Connect can automatically generate keys, standard queries, and CRUD mutations from those schemas. The example movie-review service illustrates how schemas describe fields and types, queries retrieve filtered and ordered movie data with authorization controls, and mutations insert, update, upsert, or delete records. SQL Connect is presented as a backend-as-a-service approach that combines PostgreSQL’s database capabilities with GraphQL’s structured schemas, data operations, and access controls, and it is available in public preview.

Firebase Data Connect, now renamed SQL Connect, is a Firebase service announced at Google I/O 2024 that enables applications to connect directly to PostgreSQL databases hosted on Cloud SQL. The “Everything Data Connect” blog series is intended for developers already familiar with application development and relational databases, and will explore features and best practices involving data modeling, schemas, queries, mutations, generated SDKs, end-to-end setups, relationships, pricing, transactions, aggregate fields, framework bindings, and server-side value expressions. SQL Connect is available in public preview, with documentation, sample web and Android applications, and a web codelab available to help developers get started.

Firebase’s generative AI integration, introduced in public preview as Vertex AI in Firebase in May 2024, reached general availability with client SDKs that let Android, iOS, Flutter, and web apps call Gemini models directly without requiring a custom backend; the product was renamed Firebase AI Logic in May 2025, alongside later August 2026 terminology changes for related Vertex AI services. Backed by Google Cloud standards for security, privacy, reliability, performance, and scalability, the SDKs support multimodal Gemini models, system instructions, structured JSON output constrained by schemas, function calling, and parallel function calls for connecting model responses to app capabilities or external APIs. Developers can onboard through Firebase’s console, monitor spending through the Usage and Billing dashboard, and use Firebase App Check, Remote Config, and Cloud Storage to secure access, adjust prompts and model settings remotely, and include large multimodal files. Existing preview users are directed to migration guidance, while new users can access quickstarts, documentation, and SDK repositories for support and feedback.

Firebase announced the return of its virtual Demo Day on November 19, 2024, following positive feedback from the previous year’s event. The showcase will feature demonstrations of recently introduced Firebase technologies, including Firebase Genkit, Vertex AI in Firebase—now renamed Firebase AI Logic—Gemini in Firebase, and Firebase App Hosting, with a focus on adding AI capabilities to applications, monitoring performance, and improving user experiences. Videos will be available to watch at any pace after 1:00 p.m. EST, and participants can follow Firebase on X and LinkedIn or use #FirebaseDemoDay for updates, discussion, and possible preview content.

Firebase Data Connect, now renamed SQL Connect, entered public preview as a backend-as-a-service built on Cloud SQL for PostgreSQL, enabling developers to define application data models, queries, and mutations through GraphQL-based schemas without writing custom resolvers. It automatically generates strongly typed SDKs for web, Android, and iOS, alongside a Node.js Admin SDK for privileged server-side operations, and supports relational queries, complex conditions, vector search, and retrieval-augmented generation through integration with Vertex AI, now called Agent Platform. The service offers strict and compatible schema-validation modes to support both new applications and existing PostgreSQL databases, while security integrates Firebase Authentication, CEL-based authorization, and Firebase App Check attestation. SQL Connect extends GraphQL with additional scalar types and schema-driven indexing, and its pricing combines operation and network-egress charges with Cloud SQL costs, including preview-period trial options. Documentation, sample applications, and a web codelab are available for developers who want to evaluate the platform.