InsightAgent — Turn Any CSV Into AI-Powered Stories

The Problem

Every day, individuals and businesses collect data in spreadsheets and CSV files but lack the technical expertise to extract meaningful insights. Hiring data analysts is expensive, and most visualization tools require coding knowledge.

The Solution

InsightAgent is an agentic AI system built with JavaScript that transforms any CSV dataset into actionable insights instantly. Simply upload your file and three specialized AI agents go to work:

• Agent 1 — Data Analyst: Reads your dataset and surfaces 3 specific insights
• Agent 2 — Business Advisor: Delivers one clear actionable recommendation
• Agent 3 — Chat Agent: Lets you have a real conversation about your data

How It Works

1. User uploads a CSV file through the web interface
2. The backend parses the file and sends data to Agent 1
3. Agent 1 analyzes and returns structured insights
4. Agent 2 generates a business recommendation
5. Chart.js renders the visualization automatically
6. Agent 3 stays ready to answer follow-up questions

How I Built It

Tech Stack

• Frontend: HTML, CSS, Vanilla JavaScript
• Backend: Node.js + Express
• AI: Groq API (LLaMA 3.3 70B)
• Visualizations: Chart.js
• File handling: Multer

Architecture
CSV Upload → Parse → Agent 1 → Agent 2 → UI Render → Agent 3 on standby

Responsible AI Design

• No user data stored — files deleted immediately after analysis
• All AI outputs clearly labeled as AI-generated
• Agents only reference actual data values

What I Learned

Building InsightAgent taught me how to design multi-agent pipelines where each agent has a clear bounded responsibility. Chaining agents sequentially creates emergent intelligence greater than any single prompt could achieve.

Try It Yourself

• GitHub: https://github.com/NjeriCodeCraft/InsightAgent
• Built for: JavaScript AI Build-a-thon Hack — Agents for Impact Challenge