RiskSight AI

About RiskSight AI

RiskSight AI is an AI-powered dashboard designed to detect anomalies in system logs and calculate risk scores in real-time.

The inspiration behind this project came from the increasing need for proactive cybersecurity monitoring. Traditional log analysis is time-consuming and often misses subtle anomalies that can indicate potential risks. I wanted to build a lightweight, automated tool that could help developers and system administrators quickly identify risky activity.

What I learned

• How to build a FastAPI backend for file uploads and processing CSV data.
• How to create a React frontend that communicates with the backend using Axios.
• Basic anomaly detection logic based on thresholds for login attempts and response times.
• Handling CORS issues, file uploads, and asynchronous programming in Python.
• How to structure a full-stack project ready for real-world deployment.

How I built it

• Backend: FastAPI, Python, CSV parsing, risk score calculation.
• Frontend: React, Axios, dynamic tables, and risk level display.
• Integration: Frontend uploads CSV → backend processes → returns anomalies and risk score → frontend displays results.
• Testing: Used multiple CSV files with varying data to validate anomalies detection and risk calculation.

Challenges faced

• Fixing CORS errors when React tried to call FastAPI backend.
• Handling file uploads correctly and ensuring proper CSV format.
• Displaying results dynamically in React with proper table formatting and risk-level indicators.

Built With

• axios
• css
• css-**data-format:**-csv-logs-**libraries:**-react-csv-(optional-for-csv-export)-**deployment-(optional):**-can-run-locally
• csvlogs
• fastapi
• javascript
• python
• react
• react-csv