Production Database Guides, Written for Engineers.
Skip the theory. Each playbook covers a specific problem, the TiDB features that fix it, and how to implement them.
TiDB Engineering Playbooks
Silence the Noisy Neighbor in Multi-Tenant MySQL
One bad tenant shouldn't take down the rest. This covers how to detect the source, cap it with TiDB Resource Groups, and pin repeat offenders to their own nodes with Placement Rules.
Embed a Vector DB and Build RAG Faster with TiDB
Running MySQL for transactions and a separate vector store for embeddings is two systems to babysit. TiDB handles both. This covers the native HNSW vector index, how to query embeddings alongside SQL in one statement, and a working RAG pipeline end to end.
Escape MySQL Sharding Pain and Scale Fast
Manual sharding creates problems faster than it solves them. This covers what it actually costs to maintain sharded MySQL, how TiDB handles distribution at the storage layer, and how to migrate without a maintenance window.
Build a Vibe Coding Stack That Holds Up in Production
AI-generated code ships fast and breaks in predictable ways. This covers the full stack: Next.js and TypeScript guardrails, how to set up AI coding agents with repo rules, and why TiDB Cloud handles your database, vector search, and RAG pipeline in one place — so you scale without a rewrite.
Connect Cursor to a Live Database Without Breaking Things
AI coding assistants hallucinate schema. MCP fixes that by giving Cursor live access to your tables, types, and relationships before it generates a query. This covers the full setup path for TiDB, the trust model, and what to review before any generated SQL touches production.
From Idea to Live App in a Weekend
AI coding assistants can generate an app quickly, but keeping it reliable as usage grows takes planning. This playbook covers how to scope, build, and deploy a live AI app in one weekend, with practical guidance on tool calling, RAG, vector search, production guardrails, and using TiDB Cloud to keep application data and embeddings in one scalable database.
MySQL Schema Change at Lunch, Not 2 A.M.
Schema changes shouldn't mean midnight maintenance windows. This playbook covers how TiDB's built-in Online DDL lets you add indexes and modify billion-row tables without blocking reads and writes, eliminating external tools and reducing operational overhead.
AI Agent Memory Architecture: How to Store Long-Term Memory, State, and Retrieval Data
AI agents lose context when every session starts from scratch. This playbook covers how to design production-ready agent memory, manage short-term and long-term state, and use TiDB's SQL and vector search capabilities to store and retrieve context at scale.
Migrate from Aurora MySQL and Scale Writes Without Sharding
Aurora's single-writer limits can hold back growth. This playbook covers how to migrate from Aurora MySQL to TiDB with minimal downtime, scale reads and writes horizontally, and eliminate application-level sharding while keeping MySQL compatibility.
Build an MCP Server with Persistent State Using TiDB
Most MCP tutorials stop at simple tools that lose data on restart. This playbook covers how to build an MCP server in Python, connect it to TiDB Cloud Zero for persistent SQL state, and manage session history, tool outputs, and agent memory as you move from prototype to production.
Try It on Your Own Data
MySQL-compatible, no shard config, free tier available.