JAISEV
Computer Scientist & Aspiring Machine Learning Engineer
FEATURED
A collection of digital products and experiments crafted with precision.
AI vs Real Image Detector
This project is a Binary Image Classifier capable of distinguishing between Real Photographs and AI-Generated Images (e.g., Midjourney, Stable Diffusion). It achieves ~86% accuracy using Transfer Learning on the MobileNetV2 CNN architecture. 🚀 Features Lightweight Model: Uses MobileNetV2 for fast inference, even on CPUs. Web Interface: Includes a Gradio app (app.py) to drag-and-drop images for testing. Tunable Sensitivity: The UI features a slider to adjust the detection threshold, helping to catch subtle AI fakes. Training Script: Complete, readable code (train_model.py) to train your own version on custom datasets.
ABOUT
THE AGENT
I am currently pursuing an undergraduate degree in Computer Science at the National University of Singapore (NUS).
I recently completed the International Baccalaureate Diploma Programme (IBDP), scoring 41 points, placing me in the top 5% of students worldwide. My passion for technology started early: I topped my 80-student cohort in IGCSE Computer Science and achieved a 7 in IB Higher Level Computer Science.
I'm particularly interested in cryptocurrency investing, fascinated by the potential of blockchain technology. I also have a strong interest in AI and cybersecurity, and enjoy learning about new security challenges and solutions and the potential for AI to address these challenges.
TECHNICAL ARSENAL
Skills
Frontend Architecture
Backend & Systems
DevOps & Tools
INITIATE CONTACT
Have a project in mind? Let's build something extraordinary.
© 2026 JAISEV / System v2.0