Projects
A collection of my projects spanning computer vision, medical AI, natural language processing, and generative models. Each project represents a unique challenge and learning opportunity in the field of artificial intelligence & software engineering.
Maze Problem RL Solutions
Implementation of reinforcement learning solutions for maze path-finding problems; experimenting with different RL algorithms to solve maze navigation tasks.
Monocular Depth Estimation via Transfer Learning
Recreation of a depth estimation method from a research paper using transfer learning; implemented in PyTorch though original might have used TensorFlow.
Machine Learning Algorithms (from scratch)
Reimplemented a number of machine learning algorithms from scratch, following the book “Hands-on Machine Learning with Scikit-Learn”; good exercise in understanding fundamentals.
Deep White Noise Elimination using Recurrent and Convolutional GANs
Built a model combining recurrent and convolutional layers in GAN architectures to reduce white noise in signals/images; research / experiments oriented project.
Practical Linear Algebra Exercises (Julia)
Solutions to the exercises from *Practical Linear Algebra for Data Science: From Core Concepts to Applications* by Mike X. Cohen, implemented in Julia and Jupyter notebooks.
CLRS Algorithms Implementations
Full implementations of the algorithms from *Introduction to Algorithms (CLRS)* in Java and Python; covers sorting, graph algorithms, dynamic programming, etc.
Insurance Charge Prediction
Exploratory data analysis and predictive modeling of insurance charges using a dataset; includes visualizations and baseline models (without heavy tuning / scaling).
Heart Disease Prediction
Modeling to predict heart disease outcomes (classification) using health dataset; includes data preprocessing, exploratory analysis, and predictive modeling workflows.
Multi-CNN COVID-19 Detection
Python reproduction of a paper (Elsevier) for COVID-19 detection from X-ray images using multiple pre-trained CNNs + feature selection and a Bernoulli classifier; balanced dataset and preprocessing steps included.