A FastAPI-powered backend for real-time phishing email analysis, using HuggingFace’s bert-base-uncased-emotion model. Provides secure API endpoints for emotion detection, with a custom Matrix-themed Swagger UI. Deployed on Render for live production access.
Method | Endpoint | Description |
---|---|---|
GET | / |
Healthcheck |
GET | /docs |
Custom Swagger UI (Matrix) |
POST | /predict |
Accepts email subject & body, returns emotion + confidence |
FastAPI
– lightweight web APITransformers
–bert-base-uncased-emotion
Pydantic
– request validationUvicorn
– ASGI serverRender
– hosting the live API
As part of my cybersecurity and AI engineering journey, I wanted to combine NLP and phishing detection into a live web tool. This backend powers the AI classifier using a fine-tuned BERT emotion model to analyze email content and return real-time emotional insight.
This tool is intended for educational and authorized personal use only.