🤖 Insight AI: Generative AI Market Research Platform

Insight AI is a full-stack platform that uses Generative AI to automate qualitative market research.
It dramatically reduces the time and cost of traditional methods by conducting AI-driven interviews, generating real-time summaries, and enabling deep insight discovery through an interactive Q&A bot.

---

🎥 Watch Project Demo

🚀 Key Features

• Dynamic AI Interviews — Conducts automated, adaptive conversational interviews 24/7.
• Real-Time Summarization — Instantly generates AI-powered summaries for interviews and projects.
• RAG-Powered Q&A Bot — Ask complex questions in plain language and get context-aware answers from your research data.
• Vector Database Integration — Uses ChromaDB for efficient Retrieval-Augmented Generation (RAG).
• Accessibility Focused — Integrated Text-to-Speech (TTS) for all AI responses.
• Project Dashboard — A central hub to create, manage, and share research projects.

---

🛠️ Technology Stack

Layer	Technology
Frontend	HTML5, CSS3, Vanilla JavaScript, Tailwind CSS, Chart.js
Backend	FastAPI, LangChain, Groq (Llama 3.1), SQLAlchemy
Databases	MySQL, ChromaDB

---

🏗️ System Architecture

• Frontend (SPA) → The user-facing application for project management and AI interaction.
• Backend (FastAPI) → The core API for business logic, LLM orchestration, and database management.
• LLM Service (LangChain & Groq) → Powers all generative AI tasks, from interviews to Q&A.
• Databases (MySQL & ChromaDB) →
  • MySQL stores project and user data.
  • ChromaDB manages vector embeddings for RAG-based insights.

---

⚙️ Setup Instructions

File Structure

• insight-ai/
  • backend/
    • app/
      • __init__.py - Makes the 'app' directory a Python package
      • crud.py - Handles database create, read, update, delete operations
      • database.py - Configures the connection to the MySQL database
      • llm_service.py - Contains all logic for interacting with the LLM (LangChain, Groq)
      • main.py - The main FastAPI application file, defines API endpoints
      • models.py - Defines SQLAlchemy ORM models and Pydantic schemas
      • vector_db_service.py - Manages interactions with the ChromaDB vector database
    • .env - Stores environment variables like API keys (not in version control)
    • requirements.txt - Lists all Python dependencies for the backend
  • frontend/
    • index.html - The main project dashboard page
    • app.js - JavaScript for the main dashboard
    • interview.html - The page for conducting an AI interview
    • chat.js - JavaScript for the interview chat interface
    • qa_bot.html - The page for the question-answering bot
    • qa_bot.js - JavaScript for the Q&A bot interface
    • style.css - Shared CSS for all frontend pages
  • README.md - The main project documentation file for your repository

1️⃣ Backend Setup

Clone the repository and navigate to the backend

git clone https://github.com/your-username/insight-ai.git cd insight-ai/backend

Create and activate a virtual environment

python -m venv venv

On Windows

venv\Scripts\activate

On macOS/Linux

source venv/bin/activate

Install dependencies

pip install -r requirements.txt

Create a .env file in backend/app with your Groq API key:

GROQ_API_KEY="your_api_key_here"

Run the FastAPI server:

uvicorn app.main:app --reload

📖 Usage Guide

1. Create Project — Define your research goals on the dashboard.
2. Share Link — Distribute the generated interview link to participants.
3. Monitor Progress — Track completed interviews and AI summaries in real-time.
4. Analyze Insights — Use the Q&A bot to explore your qualitative data.
5. Sync Summaries — Push processed summaries to the vector DB for RAG queries.
6. Ask Questions — Query your research data conversationally and get instant, insightful answers.