Inspiration
I wanted to make data analysis easier and faster. Large datasets can take hours to understand, so we built DataPilot, an AI agent that works in the background, finds important patterns and problems, and gives simple insights automatically.
What it does
DataPilot lets users upload a CSV dataset and automatically analyzes it in the background. It finds missing data, duplicates, outliers, and patterns, then uses Gemini to generate a simple report with useful insights and recommendations.
How we built it
I built DataPilot using Python, FastAPI, Pandas, and Gemini API. Gemini helps the agent understand the analysis and generate useful insights. We also used AI-assisted vibe coding to speed up development, debug issues, and improve the project while testing it step by step.
Challenges we ran into
The main challenge was making the analysis run in the background without blocking the app. We also had some issues with API setup, large CSV files, and connecting the Gemini API with our Python backend. AI-assisted coding helped us debug and solve these problems faster.
Accomplishments that we're proud of
I built a working AI agent that can analyze a CSV in the background, find data issues and patterns, and generate a simple report using Gemini. We are proud that we turned a simple idea into a working project.
What we learned
I learned how to connect Python, FastAPI, Pandas, and Gemini to create an agentic workflow. We also learned a lot about background tasks, API integration, and using AI-assisted coding to build faster.
What's next for DataPilot — Autonomous AI Dataset Analyst
Next, I want to support more file types, add better charts and reports, handle even larger datasets, and make the agent smarter at choosing what to investigate. We also want to add scheduled analysis so DataPilot can monitor datasets automatically.
Log in or sign up for Devpost to join the conversation.