Hybrid A* path planning + DQN reinforcement learning system in Unity ML-Agents — trains a fighter jet to navigate obstacle mazes and evade missile threats using real-time behavioral switching and a…

A clean, from-scratch PyTorch implementation of I-JEPA trained on STL-10. Built to benchmark representation-space vs. pixel-space prediction against MAE.

Real-time traffic accident detection via cross-resolution knowledge distillation from Foundation Models (DINOv2) to Edge Devices (MobileNetV3).

A PyTorch implementation of Masked Autoencoders (MAE) on the STL-10 dataset, demonstrating how self-supervised learning significantly improves classification accuracy and convergence speed on limit…

An autonomous agentic framework that generates, tests, and refines Python code, benchmarked against human developers using industry-standard quality metrics like Pylint and Maintainability Index.

An EKF-enhanced RPL routing protocol implementation in OMNeT++ for high-mobility IoT and robotic swarms.

Small part of dotcraft game

A demo unity project for Stackydash mobile game