We're deprecating Tinybird Code. Not because we've lost faith in AI — because we've learned enough to do it better.

Tinybird is a real-time data analytics platform for building fast and scalable data pipelines, offering tools for data ingestion, processing, and visualization. From this blog users can learn about stream processing, data warehousing, and analytics techniques to unlock insights from their data and make data-driven decisions.

Tinybird

Tinybird is deprecating Tinybird Code, its AI-powered CLI coding agent, not because AI failed but because the landscape shifted. The team walked through three phases of AI tooling — simple AI-assisted commands, conversational data exploration, and a full agentic CLI — before realizing that modern coding agents like Claude Code and Cursor no longer need custom wrappers. Instead, they need well-designed primitives: clear CLIs, typed SDKs, structured error messages, and good documentation. The key lesson: don't build a custom agent, make your platform work with all agents. Tinybird is replacing the agent with portable 'Agent Skills' — structured rules and best practices any coding agent can load — combined with a TypeScript SDK, CLI, and MCP server. Building the agent was still valuable because it forced improvements to error messages, API design, and overall developer experience.

We built an AI coding agent. Here's why we're sunsetting it.

The TypeScript SDK and CLI as agent interfaces