datavane/tis

Support agile Ontology DataOps Based on Flink, DataX and Flink-CDC with Web-UI

What it solves

TIS is an AI-native data integration platform designed to simplify the complex process of data engineering. It removes the need for manual, complex configurations and SQL writing by allowing users to create data pipelines and perform data analysis using natural language.

How it works

TIS combines a production-proven batch and stream processing engine (using DataX and Flink-CDC) with a modern AI layer:

  • Pipeline AI Agent: Uses LLMs to translate natural language business requirements into automated plugin selection, parameter generation, and task orchestration.
  • ChatBI: Employs a combination of an Ontology layer (defining business entities and semantics) and GraphRAG (via Neo4j) to retrieve relevant data and generate SQL queries from natural language questions.
  • DataOps Approach: Models execution flows to provide a "fool-proof" operational experience, supporting various deployment modes like Docker and Kubernetes.

Who it’s for

It is built for data engineers who want to automate pipeline creation and business users who need to gain data insights without knowing SQL.

Highlights

  • Natural Language Pipelines: Create data pipelines by simply describing the business need.
  • GraphRAG-powered ChatBI: Uses knowledge graphs to ensure precise semantic understanding and data retrieval for business queries.
  • Multi-LLM Support: Integrates with various models including DeepSeek, Tongyi Qianwen, and GPT-4.
  • Unified Engine: Supports both batch and real-time data synchronization (e.g., MySQL to StarRocks or Doris).
  • Extensible Architecture: Features a plugin-based system and micro-frontend framework for easy customization.

Related

  • Project
    DeepInsight-AI/DeepBI

    An AI-native data analysis platform that uses LLMs to allow users to explore, query, and visualize data from multiple sources via natural language conversations.

  • Project
    apconw/Aix-DB

    A conversational data analysis system (ChatBI) that uses LLMs and RAG to convert natural language queries into SQL and visual charts across multiple database types.

  • Project
    typedef-ai/fenic

    A semantic DataFrame engine that integrates LLM operators into a typed query model to turn unstructured data into reusable, inspectable pipelines.

  • Project
    Farzad-R/LLM-Zero-to-Hundred

    A collection of LLM projects and tutorials covering RAG, agentic workflows, multimodal chatbots, and fine-tuning pipelines to move from basic to advanced AI application development.

  • Project
    dataease/SQLBot

    An intelligent data questioning system based on LLMs and RAG that enables conversational data analysis (ChatBI) by converting natural language into SQL queries.