Portfolio

Outcomes you can measure

Representative work illustrating how we pair models with engineering discipline—monitoring, evals, and operator UX included.

Deepfake Face Detection

An AI-powered system that detects manipulated or synthetic human faces in images and videos using deep learning techniques.

A robust deep learning-based solution that analyzes facial features and patterns to identify deepfake content, helping prevent misinformation and digital fraud.

Tech

PythonPyTorchOpenCVCNNFastAPIReact

Problem

The rise of deepfake technology has made it difficult to distinguish real from fake media, leading to misinformation, identity misuse, and security risks.

Solution

Built a CNN-based model trained on real vs fake datasets, using image preprocessing and feature extraction. Integrated the model into a FastAPI backend with a simple UI for uploading and classifying media in real-time.