I am an ambitious Computer Science and Engineering undergraduate with a strong passion for software engineering and machine learning. I focus on building expertise in modern technologies to develop scalable, impactful solutions.
I enjoy tackling challenges, staying focused under pressure, and exploring innovative ideas to contribute effectively to projects. I currently work as a Software Engineer and apply my expertise across both software engineering and machine learning, bringing equal enthusiasm and depth to both domains.
B.Sc. Engineering (Hons) — Computer Science & Engineering
GCE Advanced Level — Physical Science Stream
Software Engineer (Part-Time) Remote
Software Engineering Intern Remote
NumPyExperienced
PandasExperienced
HuggingFaceIntermediate
LangChainIntermediate
A graph-based evaluation pipeline to benchmark reasoning capabilities and factual consistency of LLMs. Extracts structured Knowledge Graphs from LLM outputs to reduce hallucinations. In collaboration with Miami University.
An autonomous AI agent for financial market analysis and automated trading insights, implementing MCP and RAG-based knowledge retrieval using Pinecone. Features asynchronous data scraping pipelines for real-time market data.
AI-powered job marketplace for day workers in Sri Lanka. Built a RAG-based conversational recommendation system; implemented secure authentication, OTP verification, and a scalable database architecture.
Robust exchange simulator in C++ with a price-time priority order matching engine, partial fills, and order book updates. Multithreaded client-server architecture enabling concurrent multi-user trading with session persistence.
Comprehensive HR management system supporting four organisational user levels. Led frontend development with an intuitive UI and contributed to database design and implementation.
Full interpreter for the RPAL programming language: lexical analyzer, parser, AST/ST generation, and a CSE machine for program execution. Built entirely in Java.
A machine learning web app that predicts a song's genre from its lyrics. Trained a Naive Bayes classifier on 28,000+ songs across 8 genres using Apache Spark MLlib with a TF-IDF pipeline (unigrams & bigrams). Paste any lyrics to get an instant genre prediction with a probability breakdown chart.
Designed a 4-bit nano processor capable of arithmetic and logic operations, optimised for low energy consumption. Implemented entirely using digital logic design.
Open to software engineering and machine learning engineering opportunities. Feel free to reach out for projects, collaborations, or just a conversation about tech.
chamathg.21@cse.mrt.ac.lk +94 77 686 0337 chamathgunapala.me