Skip to content

MD-NAVED/Analyst.AI

Repository files navigation

Analyst.Ai Icon

Analyst.Ai

AI-Powered Data Analytics Platform

Upload. Clean. Visualize. Chat with your data β€” all in one premium app.

Live Demo React TypeScript Vite Gemini AI License


🌟 What is Analyst.Ai?

Analyst.Ai is a premium, browser-based data analytics suite that turns raw CSV/Excel files into actionable insights in seconds. Powered by Google Gemini AI β€” with smart cleaning, dynamic chart generation, natural language querying, and pixel-perfect PDF export β€” it's everything a data analyst needs in one sleek interface.

No Python. No Jupyter. No setup. Just upload and explore.

πŸ”— Try it live: data-lens-ai-omega.vercel.app


✨ Feature Highlights

πŸ“ Smart Data Ingestion

  • Supports CSV, Excel (.xlsx/.xls), JSON, TSV, Google Sheets, MySQL & PostgreSQL
  • High-performance chunk-based parsing with real-time progress bar and ETA for files up to 500,000 rows
  • Auto-detects column types: number, text, date
  • Auto-configurable CSV delimiters (comma, tab, semicolon, pipe)

πŸ€– AI-Powered Features (Gemini)

  • AI Chat β€” Ask natural language questions about your dataset ("What is the average revenue by region?")
  • Dynamic Chart Generation β€” AI writes JavaScript transformation code and renders live interactive charts
  • Auto-Analyst β€” One-click full analysis: cleaning β†’ statistics β†’ insights β†’ charts β†’ report
  • EDA Report β€” Auto-generated exploratory data analysis with AI-written insights
  • Smart Chart Recommendations β€” AI analyzes your data and recommends the most relevant chart types automatically

🧹 Data Cleaning

  • Drop null rows / fill with mean, median, mode, or custom value
  • Remove duplicate rows
  • Convert column data types (text ↔ number ↔ date)
  • Drop unwanted columns
  • Full undo/redo history for every cleaning operation

πŸ“Š Smart Visualizations

  • AI-generated charts β€” dynamically rendered from natural language queries
  • Chart Picker Bar β€” pill-style toggles to add/remove charts; max 4 active at a time
  • Supported chart types: Bar, Area, Line, Pie, Scatter Plot, Histogram
  • Full-screen Zoom Mode with interactive Brush slider
  • PNG & SVG download for every chart

πŸ“ˆ Statistics & EDA

  • Per-column statistics: mean, median, std dev, min/max, nulls, unique values
  • Distribution preview per column
  • Correlation insights
  • AI-generated written insights for the entire dataset

πŸ—‚οΈ Dashboard Builder

  • Drag-and-drop widget canvas (powered by react-grid-layout)
  • Widget types: KPI cards, Line charts, Bar charts, Tables, Markdown text
  • Dashboards auto-saved per dataset to localStorage
  • Print/PDF export of entire dashboard

πŸ“„ PDF Export

  • Multi-page executive PDF reports via jsPDF + html2canvas
  • Auto-avoiding page breaks β€” charts and titles never split mid-page
  • Dynamic layout budgeting
  • Auto-generated Table of Contents with page numbers

πŸ” Authentication & Onboarding

  • Email/password registration and login
  • Google OAuth social login
  • Multi-step onboarding flow for new users
  • Persistent sessions via IndexedDB

πŸ› οΈ Technology Stack

Layer Technology
Frontend Core React 19, TypeScript 5, Vite 6
Styling Vanilla CSS with CSS variables, Glassmorphism design system
Charts Recharts (custom tooltips, gradients, animations)
Data Parsing PapaParse (CSV), XLSX (Excel), custom JSON/TSV parser
PDF Export jsPDF + html2canvas (high-DPI, dynamic layout)
AI / LLM Google Gemini Flash/Pro via @google/genai SDK
Dashboard react-grid-layout (drag-and-drop canvas)
Storage IndexedDB (via idb) for session persistence
Backend Node.js + Express (DB import API, auth)
Cloud DB Supabase (PostgreSQL)
Icons Lucide React

πŸš€ Getting Started

Prerequisites

Setup

  1. Clone the repository:

    git clone https://github.com/MD-NAVED/Analyst.AI.git
    cd Analyst.AI
  2. Install dependencies:

    npm install
  3. Configure environment: Create a .env file in the root directory:

    VITE_GEMINI_API_KEY=your_gemini_api_key_here
  4. Start development server:

    npm run dev

    Open http://localhost:3001 in your browser.

  5. Production build:

    npm run build

πŸ“Έ App Walkthrough

Step What you get
1. Upload Drag & drop CSV/Excel/Google Sheets β€” parsed instantly
2. Preview Browse data table with search, filter, and column type badges
3. Clean One-click AI cleaning or manual operations with full undo history
4. Statistics Per-column stats with AI written insights
5. Visualize Smart chart picker with AI-recommended charts, zoom, download
6. EDA Report Full exploratory analysis with AI commentary
7. AI Chat Natural language Q&A β€” AI generates code & renders live charts
8. Dashboard Build a custom drag-and-drop KPI dashboard
9. Export Download PDF, PNG, SVG, CSV, or Excel

πŸ—οΈ Enterprise Architecture (GCP)

Analyst.Ai is designed to scale from local CSV analysis to an enterprise cloud analytics platform. Below is the serverless GCP deployment blueprint:

graph TD
    User([User Browser]) -->|HTTPS| LB[Global HTTPS Load Balancer]
    subgraph GCP_Cloud ["Google Cloud Platform"]
        LB -->|Serverless NEG| CR[Cloud Run: Analyst.Ai App]
        CR -->|Secrets Auth| SM[Secret Manager]
        CR -->|SQL Proxy| SQL[(Cloud SQL: PostgreSQL 15)]
        CR -->|Streaming| BQ[(BigQuery: Dataset)]
        CR -->|LLM Calls| Vertex[Vertex AI / Gemini]
        SQL -..->|DB Password| SM
    end
Loading
Component Role
Cloud Run Hosts the Dockerized app, auto-scales 0 β†’ 1000+ instances
Secret Manager Stores API keys, JWT secrets, DB passwords securely
Cloud SQL Managed PostgreSQL for user accounts and session data
BigQuery High-volume dataset analytics warehouse
Vertex AI Private, secure access to Gemini Pro/Ultra models
Terraform Full Infrastructure-as-Code for repeatable deployments

πŸ“„ License

This project is licensed under the MIT License β€” see the LICENSE file for details.


Made with ❀️ using React, TypeScript & Google Gemini AI

⭐ Star this repo if you find it useful!

πŸ”— Live Demo Β· πŸ› Report Bug Β· πŸ’‘ Request Feature


🌟 What is DataLens AI?

DataLens AI is a premium, browser-based data analytics suite that turns raw CSV/Excel files into actionable insights in seconds. With AI-powered cleaning, smart chart recommendations, natural language querying, and pixel-perfect PDF export β€” it's everything a data analyst needs in one sleek interface.

No Python. No Jupyter. No setup. Just upload and explore.


✨ Feature Highlights

πŸ“ Smart Data Ingestion

  • Supports CSV, Excel (.xlsx/.xls), JSON, and TSV formats
  • High-performance chunk-based parsing with real-time progress bar and ETA for files up to 500,000 rows
  • Auto-detects column types: number, text, date
  • Auto-configurable CSV delimiters (comma, tab, semicolon, pipe)

πŸ€– AI-Powered Features (Gemini)

  • AI Chat β€” Ask natural language questions about your dataset ("What is the average revenue by region?")
  • Auto-Analyst β€” One-click full analysis: cleaning β†’ statistics β†’ insights β†’ charts β†’ report
  • EDA Report β€” Auto-generated exploratory data analysis with AI written insights
  • Smart Chart Recommendations β€” AI analyzes your data and recommends the most relevant chart types automatically

🧹 Data Cleaning

  • Drop null rows / fill with mean, median, mode, or custom value
  • Remove duplicate rows
  • Convert column data types (text ↔ number ↔ date)
  • Drop unwanted columns
  • Full undo/redo history for every cleaning operation

πŸ“Š Smart Visualizations

  • AI-recommended charts β€” only renders charts relevant to your data (not all 6 at once)
  • Chart Picker Bar β€” pill-style toggles to add/remove charts; max 4 active at a time to prevent lag
  • Supported chart types: Bar, Area, Line, Pie, Scatter Plot, Histogram
  • Full-screen Zoom Mode with interactive Brush slider for every chart type including Histogram
  • PNG & SVG download for every chart
  • Sampled view mode for large datasets

πŸ“ˆ Statistics & EDA

  • Per-column statistics: mean, median, std dev, min/max, nulls, unique values
  • Distribution preview per column
  • Correlation insights
  • AI-generated written insights for the entire dataset

πŸ—‚οΈ Dashboard Builder

  • Drag-and-drop widget canvas (powered by react-grid-layout)
  • Widget types: KPI cards, Line charts, Bar charts, Tables, Markdown text
  • Dashboards auto-saved per dataset to localStorage
  • Print/PDF export of entire dashboard

πŸ“„ PDF Export

  • Multi-page executive PDF reports via jsPDF + html2canvas
  • Auto-avoiding page breaks β€” charts and titles never split mid-page
  • Dynamic layout budgeting β€” compresses chart heights to fit on one page before forcing breaks
  • Auto-generated Table of Contents with page numbers

πŸ” Authentication & Onboarding

  • Email/password registration and login
  • Google OAuth social login
  • Multi-step onboarding flow for new users
  • Persistent sessions via IndexedDB

πŸ› οΈ Technology Stack

Layer Technology
Frontend Core React 19, TypeScript 5, Vite 5
Styling Vanilla CSS with CSS variables, Glassmorphism design system
Charts Recharts (fully customized β€” custom tooltips, gradients, animations)
Data Parsing PapaParse (CSV), XLSX (Excel), custom JSON/TSV parser
PDF Export jsPDF + html2canvas (high-DPI, dynamic layout)
AI / LLM Google Gemini API via @google/genai SDK
Dashboard react-grid-layout (drag-and-drop canvas)
Storage IndexedDB (via idb) for session persistence
Icons Lucide React

πŸš€ Getting Started

Prerequisites

Setup

  1. Clone the repository:

    git clone https://github.com/MD-NAVED/DataLens-AI.git
    cd DataLens-AI
  2. Install dependencies:

    npm install
  3. Configure environment: Create a .env file in the root directory:

    VITE_GEMINI_API_KEY=your_gemini_api_key_here
  4. Start development server:

    npm run dev

    Open http://localhost:3001 in your browser.

  5. Production build:

    npm run build

πŸ“Έ App Walkthrough

Step What you get
1. Upload Drag & drop CSV/Excel β€” parsed instantly with progress bar
2. Preview Browse data table with search, filter, and column type badges
3. Clean One-click AI cleaning or manual operations with full undo history
4. Statistics Per-column stats with AI written insights
5. Visualize Smart chart picker with AI-recommended charts, zoom, download
6. EDA Report Full exploratory analysis with AI commentary
7. AI Chat Natural language Q&A on your dataset
8. Dashboard Build a custom drag-and-drop KPI dashboard
9. Export Download PDF, PNG, SVG, CSV, or Excel

πŸ—οΈ Enterprise Architecture (GCP)

DataLens AI is designed to scale from local CSV analysis to an enterprise cloud analytics platform. Below is the serverless GCP deployment blueprint:

graph TD
    User([User Browser]) -->|HTTPS| LB[Global HTTPS Load Balancer]
    subgraph GCP_Cloud ["Google Cloud Platform"]
        LB -->|Serverless NEG| CR[Cloud Run: DataLens App]
        CR -->|Secrets Auth| SM[Secret Manager]
        CR -->|SQL Proxy| SQL[(Cloud SQL: PostgreSQL 15)]
        CR -->|Streaming| BQ[(BigQuery: Dataset)]
        CR -->|LLM Calls| Vertex[Vertex AI / Gemini]
        SQL -.->|DB Password| SM
    end
Loading
Component Role
Cloud Run Hosts the Dockerized app, auto-scales 0 β†’ 1000+ instances
Secret Manager Stores API keys, JWT secrets, DB passwords securely
Cloud SQL Managed PostgreSQL for user accounts and session data
BigQuery High-volume dataset analytics warehouse
Vertex AI Private, secure access to Gemini Pro/Ultra models
Terraform Full Infrastructure-as-Code for repeatable deployments

πŸ“„ License

This project is licensed under the MIT License β€” see the LICENSE file for details.


Made with ❀️ using React, TypeScript & Google Gemini AI

⭐ Star this repo if you find it useful!

About

A powerful data analysis and visualization tool powered by the Google Gemini API to extract automated insights from CSV/Excel files.

Resources

Stars

1 star

Watchers

0 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors