Fake News Classifier
95.76% accuracy on a 72,000-article datasetTF-IDF + Logistic Regression model for classifying fake news headlines.
- Python
- scikit-learn
- NLP
Jaden Isaac
> AI Systems Builder — building from first principles
About
BTech student in AI & ML at Amity University Bangalore, entering his 5th semester.
Builds AI systems from first principles rather than wiring together frameworks — implementing backpropagation, optimizers, and inference engines from scratch to actually understand what's happening under the hood.
Featured Projects
C++ Neural Network → WebAssembly
Handwritten digit recognizer with backpropagation implemented from scratch in C++, compiled to WebAssembly and running in-browser via an interactive HTML Canvas.
- C++
- WebAssembly
- Neural Networks
NumPy Neural Network from Scratch
97.87% accuracy on MNISTFully connected neural network built with NumPy only, comparing SGD, Momentum, and Adam optimizers.
- Python
- NumPy
- Deep Learning
Rockfall Prediction System
AI-powered system predicting rockfall incidents in open-pit mines, built solo for Smart India Hackathon 2024.
- XGBoost
- FastAPI
- React
Contact
Reach out through any of the channels below.