Projects

A chess engine that demonstrates advanced algorithms through tree search optimization. [Read More]

A desktop application built with Tauri and Vue.js (TypeScript) that provides real-time speech-to-text functionality using Vosk. [Read More]

This project is a PyTorch-inspired deep learning framework built from scratch, not aimed at competing with established frameworks but to provide a deeper understanding of how deep neural networks work under the hood. The goal is to recreate the core functionality of PyTorch, including tensor operations, automatic differentiation, and basic... [Read More]

This project is not a traditional data engineering pipeline like the ones used to move data between systems. Instead, it’s a pipeline specifically designed to process data for deep learning workflows. It focuses on tasks like loading, preparing, transforming, and augmenting data right before it’s fed into a machine learning... [Read More]

In this project, I applied Denoising Diffusion Probabilistic Models (DDPM) to audio data. Diffusion models have been popular in image generation, but my focus was to adapt this method for audio processing, particularly for data augmentation and creative audio manipulation. [Read More]

This was a relatively short but fascinating project where I experimented with Vision Transformers (ViT)—a model architecture I hadn’t explored much before. The main goal was to apply a pre-trained ViT for emotion classification. I learned a lot about how ViTs work and their potential applications in computer vision tasks.... [Read More]

This was one of my first deep learning projects, and the one that sparked my love for deep learning. Working on music genre classification using a CNN was a great introduction to audio data processing and inspired me to explore more about neural networks and their applications. [Read More]