--- layout: post title: "Announcing Database AI suggestions: from table name to schema in one click" description: We’re introducing new AI features into Appwrite Databases that generate complete schemas and indexes from just a table name. date: 2025-11-28 lastUpdated: 2026-06-29 cover: /images/blog/announcing-database-ai-suggestions/cover.avif timeToRead: 5 author: darshan-pandya category: announcement featured: false faqs: - question: "How do Database AI suggestions decide which columns and indexes to create?" answer: "Suggestions are generated from the table's name (like `Orders` or `UserSessions`), the structure of sibling tables in the same database, and any extra context you provide in the prompt. [Appwrite](/docs/products/databases) uses this to propose sensible field types, naming conventions, and recommended indexes. You can review and tweak everything before applying." - question: "Do AI suggestions apply automatically when I create a table?" answer: "No, suggestions never apply automatically. You opt in by enabling Generate AI suggestions during table creation, then review the proposed schema, adjust columns inline, or skip suggestions entirely. Applying the final schema (including any suggested indexes) takes one click." - question: "Can I generate sample data alongside the schema?" answer: "Yes, there's an option during table setup to populate the table with example records that match the generated or manually defined schema. The sample entries follow your field types and naming conventions, which makes it easy to validate queries, indexes, and relationships before integrating the table into your app." - question: "Will AI suggestions overwrite an existing table's schema?" answer: "No, Database AI suggestions are only offered when you're creating a new table. Existing tables stay untouched. If you want to evolve a schema, you can add columns and indexes manually or use [Appwrite's migration APIs](/docs/products/databases)." - question: "Why use AI suggestions instead of designing the schema by hand?" answer: "Schemas designed by hand often miss timestamps, soft-delete fields, or indexes that you only realize you need later. AI suggestions bake those best practices in from the start, reduce naming inconsistencies across tables, and turn a slow setup step into a few seconds of review." - question: "Is Database AI Suggestions available on self-hosted Appwrite?" answer: "Database AI Suggestions launched first on [Appwrite Cloud](/docs/products/databases). Check the Appwrite changelog for self-hosted availability, since features that depend on AI services are typically rolled out to Cloud before being available in self-hosted releases." --- In many development workflows, setting up a database schema is one of the first and often one of the slowest steps. Starting with a blank schema, juggling naming conventions, remembering which fields need indexes, or ensuring every table follows best practices can quickly become a chore. Whether spinning up a new project, designing an operational table for a feature, or standardizing schemas across multiple services, the setup should be fast, consistent, and reliable. That’s why we’re introducing **Database AI suggestions.** A simple but powerful way to generate sensible columns and indexes the moment you create a new table. By analyzing the table name, sibling tables, and any extra context you provide, Appwrite suggests a complete schema that you can tweak and apply in seconds. No more blank slates, no more forgotten indexes, and no more inconsistent naming. # Design schemas in seconds, not hours When creating a new table in the Appwrite Console, you can **choose to generate AI-powered suggestions** for columns and indexes tailored to your use case. For example, if you name your table `Orders`, Appwrite will suggest fields like `orderId`, `userId`, `totalAmount`, and `status`, complete with sensible types and recommended indexes. These suggestions are generated using a mix of: - The table’s **name** (e.g., “Orders”, “UserSessions”) - **Sibling tables and their respective columns** in your database to align naming and structure - Any **additional context** you provide in your prompt You’re always in control. Suggestions don’t apply automatically. You can review them, adjust inline, or skip them entirely. Once ready, applying the schema is just one click, including an optional CTA to add suggested indexes for even better performance. Alongside Database AI Suggestions, we have also added an **AI-powered index suggestions UI.** This UI update complements Database AI Suggestions by making the schema design and feedback process more guided, interactive, and flexible across both desktop and mobile. # Immediate benefits **Database AI Suggestions** isn’t just about saving a few clicks. It removes friction from a critical early step in your workflow and sets you up for success. For example, when setting up schemas daily, managing operational data, and standardizing schemas across microservices, this new feature gives you a smart starting point, helping teams of all sizes move faster while maintaining quality and consistency. Resulting in benefits such as: - **Faster setup:** Go from table name to sensible schema in seconds. - **Consistency:** Keep field names and types aligned across tables. - **Best practices baked in:** Get timestamps, soft-delete fields, enums, and indexes without having to remember them. - **Fewer errors:** Avoid schema issues that cause analytics problems or require migrations later. # Generate sample data for faster validation We’ve also added an option to quickly generate sample data when creating a new table. This button appears during table setup and populates the table with example records that match the generated (or manually defined) schema. The sample entries follow the field types and naming conventions you’ve chosen, making it easier to validate assumptions early, test queries, and confirm that indexes and relationships behave as expected. This is especially useful when prototyping or verifying a schema before integrating it into your application. # Get started Database AI Suggestions is now live on Appwrite Cloud. Create a new table and enable “Generate AI suggestions” to view proposed columns, indexes, and optional sample data. # More resources - [Read the documentation](/docs/products/databases/ai-suggestions) - [Appwrite MCP server](/docs/tooling/mcp) - [Make the best use of Appwrite’s MCP server](/blog/post/make-best-use-appwrite-mcp) - [Announcing an improved Appwrite Databases experience. A completely new look and feel](/blog/post/announcing-appwrite-databases-new-ui)