February 2025 Summaries

Firebase Data Connect, now renamed SQL Connect, is a Cloud SQL-based backend-as-a-service for PostgreSQL that provides scalable, secure relational-data access through GraphQL. Recent updates add generated React SDKs with TanStack Query bindings, PGLite-based emulator data import and export, vector similarity search distance metadata to help tune queries, and expanded query authorization through CEL-powered `@check` rules. The release also introduces an Android quickstart application and links to a livestream demonstration, while planned work includes Angular bindings, broader CEL support, and COUNT query functionality. The team encourages user feedback through Firebase UserVoice as it continues developing the platform in 2025.

Google has made Gemini Code Assist available at no cost for individual developers, offering up to 6,000 code completions and 240 chat interactions per day to support Firebase application development. Complementing Gemini in Firebase, which provides console-based help with SDK setup, Firebase-specific code generation and optimization, troubleshooting, crash analysis, and GraphQL queries, the IDE extension uses Gemini 2.0 and local codebase context to generate code, complete code, and answer questions. It supports Visual Studio Code, JetBrains IDEs, Cloud Workstations, and Cloud Shell Editor, along with languages including Java, Kotlin, Swift, JavaScript, Dart, C++, C#, Python, and Go. Gemini Code Assist for GitHub also reviews pull requests for bugs and style issues, while paid Standard and Enterprise plans provide advanced capabilities such as productivity metrics, private-code-based responses, and Google Cloud integrations.

Google’s Gemini 2.0 Flash, used through the Genkit AI integration framework for Node.js, can analyze PDF documents multimodally and extract information without a separate OCR system. The sample command-line application demonstrates extracting invoice fields and line items from a PDF, then applying Zod schemas to return predictable, machine-processable structured output such as invoice numbers, dates, quantities, and totals. Genkit’s unified generation API can also support models without native structured-output capabilities by augmenting prompts and coercing results. For reusable workflows across different document types, the post uses Genkit’s Dotprompt files to separate prompts, model configuration, input parameters, and output schemas from application code, enabling the same extraction function to process invoices, handwritten forms, and other PDFs by selecting an appropriate prompt. The author highlights Gemini’s ability to interpret handwritten entries and points readers to Genkit documentation, a GitHub sample, and related livestream projects.

Firebase Genkit for Node.js has reached version 1.0 and is now production-ready, providing an open-source, developer-focused framework for adding AI agents, automations, and generative features to applications. It supports Google Gemini, Imagen, Anthropic via Vertex AI, community integrations such as OpenAI and GitHub, and self-hosted models including Gemma, Llama, and DeepSeek through Ollama, while its Go version remains in alpha. Genkit offers unified and structured generation APIs, function calling, vector search for retrieval-augmented generation, and local tools including a CLI and Developer UI for prompt experimentation, debugging, tracing, and evaluation. Deployments on Firebase or Google Cloud can use built-in monitoring dashboards to examine token usage, latency, failures, and production traces. Version 1.0 also stabilizes core capabilities for agentic applications, while beta features provide multi-turn chat, session state, and multi-agent workflows; documentation, examples, migration guidance, and community feedback channels are available for developers getting started.

Firebase Data Connect, now renamed SQL Connect, uses GraphQL schemas to model structured, persistent application data, with the movie review app illustrating how types, tables, columns, and directives define database structure. Schemas use the `@table` directive to create tables and optionally customize table names, singular and plural operation names, and key fields, while field definitions specify supported data types, required values, database column details, default expressions, and indexes. Defaults reduce configuration needs by automatically naming columns and keys, while directives such as `@col`, `@default`, and `@index` allow developers to customize SQL column types, generate values such as UUIDs or request timestamps, and improve query performance through single-field or composite indexes. SQL Connect also creates key scalars from schema key fields to support record identification and relational operations, provides server-computed values for fields generated at creation time, and supports one-to-one, one-to-many, and many-to-many table relationships for relational data modeling.