Projects
Projects
A bunch of small but fun data science projects where I try out new ideas and tools. They’re deliberately lightweight and exploratory, focused on learning by doing rather than building anything heavy or production-ready. Each one is a quick end-to-end experiment to test a concept, see what works, and pick up something useful along the way.
A Retrieval Augmented Generation (RAG) LLM
Built a simple Retrieval Augmented Generation (RAG) toolkit API using OpenAI and FastAPI to facilitate LLM integrations with a curated and vectorised knowledge base.
Transfer Learning for Trash Detection with YOLOv11
Applied transfer learning concepts to a YOLOv11 model for accurate real-time trash detection, using macbook webcam for local demonstration and testing purposes.
DSP and ML for Arrhythmia Classification
Developed an advanced machine learning pipeline with digital signal processing techniques to identify and classify anomalous heartbeats into various types of arhythmia, for the development of improved ICDs
Interactive Digit Machine Learning
Interactive CNN-based digit classifier web app using Gradio, allowing users to draw digits and receive real-time predictions with visual explanations. Project explores the use of CICD to test and deploy updated/retrained models.
Interested in collaborating?
I'm always open to discussing new projects and opportunities.
GET IN TOUCH