RocketRide
Self-hosted AI pipeline builder with a VS Code canvas, Python and TypeScript SDKs, and an MIT-licensed engine that can run locally or in Docker.
RocketRide is a self-hosted AI pipeline builder and runtime for developers building document search, data processing, and AI agent workflows. Its visual editor lives in VS Code, so you can design and debug workflows alongside your application code. The engine is free and open source under the MIT license.
You can run it on your own machine, in Docker, or on an on-premises server. Local use doesn't require an account, and deployments can run fully air-gapped with local models and services. The included document-processing example extracts text from PDFs and images and removes personal data without sending it off your machine. Connections to OpenAI or Anthropic use external model services; RocketRide Cloud hosts the engine for you.
Pipelines combine model calls with document parsing, embeddings, vector search, and tools. Qdrant supports local document retrieval, while CrewAI and LangChain support agent workflows with memory shared across runs. Developers can extend pipeline nodes in Python or expose pipelines as tools for AI assistants through MCP.
The visual editor saves workflows as portable JSON files that you can review and version in Git. The same pipeline can move between your own infrastructure and the hosted service without a rewrite. A multithreaded C++ engine executes the workflows, and Python and TypeScript SDKs connect them to applications. Execution traces show model calls, token usage, latency, and memory consumption for debugging and performance analysis.
