Analytics AI Agent is an AI-powered data visualization system that enables users to upload structured datasets (CSV/Excel) and generate insights using natural language queries.
Instead of manually writing aggregation logic or configuring charts, users can simply ask:
- “Show sales trend by region over time.”
- “Compare revenue distribution across categories.”
The system interprets the query using LLM-based reasoning and dynamically generates relevant visualizations.
Traditional data exploration requires:
- SQL knowledge
- BI tool expertise
- Manual configuration of visualizations
This project removes that friction by enabling:
- Conversational analytics
- Automated chart generation
- Faster insight extraction
It bridges the gap between raw datasets and actionable insights.
- Supports CSV and Excel files
- Automatic parsing and schema detection
- Natural language understanding via LLM integration
- Converts queries into structured data operations
- Dynamically selects suitable visualization types
- Suggested visualizations
- Responsive UI
- Clean component-based structure
- Export visualizations as images
- Useful for reports and presentations
- React (TypeScript)
- Vite
- Tailwind CSS
- shadcn-ui
- LLM service abstraction
- Query-to-visualization logic generation
- Custom React hooks for dataset handling
- Utility modules for transformation and aggregation
├── public
├── src
│ ├── components
│ │ ├── ui
│ │ ├── ChatInterface.tsx
│ │ ├── Dashboard.tsx
│ │ ├── DataVisualizer.tsx
│ ├── hooks
│ │ └── useDataProcessing.ts
│ ├── lib
│ │ └── dataUtils.ts
│ ├── pages
│ │ └── Index.tsx
│ ├── services
│ │ ├── groqService.ts
│ │ └── llmService.ts
│ ├── App.tsx
│ ├── main.tsx
├── package.json
└── tailwind.config.ts
- Node.js (v18+ recommended)
- npm
git clone https://github.com/sanskaryo/Analytics-AI-Agent.git
cd Analytics-AI-Agent
npm install
npm run devApplication runs at:
http://localhost:5173
- Business performance analysis
- Academic dataset exploration
- Startup KPI dashboards
- Rapid exploratory data analysis (EDA)
- Internal analytics tools
- Modular service abstraction for easy LLM swapping
- Separation of UI and processing logic
- Type-safe architecture using TypeScript
- Scalable component structure
- Multi-dataset comparison
- SQL query generation mode
- Advanced statistical summaries
- Persistent query history
- Role-based access control
Supports deployment on:
- Vercel
- Netlify
Build command:
npm run buildMIT License
Sanskar Khandelwal
