AI Threat Detector

Inspiration

Small and mid-sized organizations often lack access to enterprise-level tools for monitoring and managing security threats. We wanted to create a simple, user-friendly threat detection platform that can help teams stay alert, track incidents, and respond quickly without needing a full-scale Security Operations Center (SOC).

What it does

Threat Detector is a lightweight, real-time threat monitoring and alerting system. It:

Displays live threat data on an interactive dashboard Allows teams to log, track, and prioritize threats Supports manual alert tagging, resolution tracking, and historical analysis Enables background processing of threat feeds (e.g., logs, external sources) Provides a centralized platform for managing security incidents

How we built it

We structured the system into three components:

Frontend Built with React, TypeScript, and Tailwind CSS Provides a clean dashboard interface with data visualizations and threat logs Backend Developed using Django REST Framework Handles APIs, user authentication, and threat data management Includes Celery + Redis for background task handling (e.g., parsing feeds, sending alerts) Feed Ingestion & Parsing Simple background job system that ingests sample logs (JSON/CSV format) Extracts key information and pushes it into the backend for dashboard display.

Challenges we ran into

Getting background tasks (Celery + Redis) to run smoothly in sync with the Django server Building reusable components in the frontend that are scalable and responsive Designing a clean, intuitive UI that doesn’t overwhelm the user with too much data Avoiding over-engineering—keeping it lightweight and manageable without AI/ML

Accomplishments that we're proud of

Successfully integrating a full-stack system with real-time updates Building a modular codebase that can easily be extended (e.g., with ML later) Creating a polished UI for quick understanding and management of threats Making it easy to deploy locally or via Docker

What we learned

How to coordinate frontend/backend communication in a full-stack web app The importance of clean UX in platforms where clarity is crucial (like threat dashboards) How to manage task queues and background processing in a Django environment The value of starting simple and adding features progressively

What's next for AI Threat Detector

Log format flexibility – add support for XML, syslog, etc. Scheduled reports via email (daily/weekly summaries) Custom alert rules for different severity levels Role-based access control (admin vs. analyst views) Optional cloud deployment (e.g., Render, Heroku, or Docker Swarm)

Built With

• css
• javascript
• python
• react