AI

DolphinMind is an enterprise-grade RAG question-answering system built on DolphinDB's powerful vector engine, text engine, and computing capabilities. It is designed to answer a wide range of questions that arise when using DolphinDB and DLang in daily work. It uses a hybrid retrieval, multi-channel recall, and reranking architecture, deeply integrating the DolphinDB knowledge base with LLM capabilities to deliver extremely fast responses and logically precise answers for professional Q&A while eliminating LLM hallucinations.

DolphinMind provides a web interface that allows you to query DolphinDB-related questions at any time, eliminating the cumbersome process of looking up documentation manually. DolphinMind also supports integration with AI agents such as Codex and Claude Code. By adding retrieval capabilities for the DolphinDB knowledge base, it improves the success rate of DolphinDB-related tasks executed by agents.

Starting from version 3.00.6, DolphinDB supports the agent framework DolphinX. DolphinX provides capabilities such as session management, context management, model integration, skill management, MCP management, memory management, secure execution, and permission control. It helps organizations quickly build DolphinDB-based agents and implement unified operations, governance, and O&M. You can use DolphinX in the following ways:

• Use built-in agents or user-defined agents on the web interface to interact with DolphinDB through natural language and perform tasks such as database operations, script writing, script execution, result interpretation, and troubleshooting.
• Call the DolphinX API to integrate DolphinX capabilities into enterprise-grade agents.

Starting from version 3.00.4, DolphinDB supports the MCP Server. You can develop MCP tools and connect the MCP Server to AI agents, providing agents with DolphinDB data analysis capabilities.

Starfish AI is an intelligent investment research platform built on DolphinDB. By introducing cutting-edge AI technologies, it upgrades traditional factor development, strategy generation, and data analysis workflows from manual coding to intelligent automation, significantly improving the intelligence level and research efficiency of financial institutions in quantitative research.