Hashara Kumarasinghe

Hashara Kumarasinghe

PhD Candidate · Computational Phylogenomics Lab, School of Computing · The Australian National University

I work on computational phylogenetics — applying machine learning and high-performance computing to phylogenetic tree inference.

My current project is GPU-accelerated maximum likelihood phylogenetics in IQ-TREE 3, making large-scale evolutionary analysis practical on modern HPC clusters. It is supported by a TALO Computational Biology Innovator Grant.

Before starting the PhD I was a software engineer at Sysco LABS, working on cloud infrastructure with AWS, Terraform, Progress Chef and Jenkins, and contributing to an AIX-to-AWS migration covering more than 100 systems. I hold a BSc in Computer Science and Engineering from the University of Moratuwa.

Hashara Kumarasinghe

Research

My doctoral research develops machine learning and high-performance computing methods for phylogenetic tree inference, in the Computational Phylogenomics Lab at ANU. Two threads run through it: using deep learning to make phylogenetic model selection faster, and implementing GPU-accelerated maximum likelihood estimation in IQ-TREE 3 so that large phylogenomic datasets can be analysed on modern HPC clusters.

Earlier work, during my undergraduate degree at the University of Moratuwa, applied deep learning to chest radiography for pneumonia and Covid-19 identification.

290 citations · h-index 5 · i10-index 4 — Google Scholar, August 2026

Preprints

  1. 2025

    IQ2MC: A New Framework to Infer Phylogenetic Time Trees Using IQ-TREE 3 and MCMCTree with Mixture Models

    P. Demotte, M. Panchaksaram, H. Kumarasinghe, N. Ly-Trong, M. dos Reis, B. Q. Minh

    EcoEvoRxiv

Journal articles

  1. 2022

    Chest X-ray analysis empowered with deep learning: A systematic review

    D. Meedeniya, H. Kumarasinghe, S. Kolonne, C. Fernando, I. de la Torre Díez, G. Marques

    Applied Soft Computing, vol. 126, art. 109319

  2. 2022

    U-Net Based Chest X-ray Segmentation with Ensemble Classification for Covid-19 and Pneumonia

    K. A. S. H. Kumarasinghe, S. L. Kolonne, K. C. M. Fernando, D. Meedeniya

    International Journal of Online and Biomedical Engineering (iJOE), vol. 18, no. 7, pp. 161–175

Conference papers

  1. 2022

    Chest Radiographs Classification Using Multi-model Deep Learning: A Comparative Study

    C. Fernando, S. Kolonne, H. Kumarasinghe, D. Meedeniya

    2nd International Conference on Advanced Research in Computing (ICARC), pp. 165–170

  2. 2021

    MobileNetV2 Based Chest X-Rays Classification

    S. Kolonne, C. Fernando, H. Kumarasinghe, D. Meedeniya

    International Conference on Decision Aid Sciences and Application (DASA), pp. 57–61

Posters

  1. 2025

    Hybrid Phylogenetic Model Selection using Deep Learning and High-Performance Computing

    H. Kumarasinghe, T. Drucks, T. Wong, A. von Haeseler, B. Q. Minh

    AJCAI 2025 Workshop on AI/ML in Computational and Experimental Biology

  2. 2025

    Poster presentation

    H. Kumarasinghe

    ABACBS 2025 — Australian Bioinformatics and Computational Biology Society, Adelaide

Peer review

  1. 2025

    Systematic Biology

    Pre-publication reviewer · Oxford University Press

  2. 2024

    Journal of Computational Biology

    Pre-publication reviewer · Mary Ann Liebert / SAGE

Verified review record on ORCID.

Projects

Current work

  1. 2026 –

    GPU-accelerated maximum likelihood phylogenetics in IQ-TREE 3

    PhD project · Computational Phylogenomics Lab, ANU

    Maximum likelihood tree inference spends most of its runtime evaluating the likelihood of candidate trees, and that cost grows quickly with the size of the alignment. This project moves the likelihood calculation onto GPUs in IQ-TREE 3, with the aim of making genome-scale phylogenetic analyses practical on modern HPC clusters.

    Supported by a TALO Computational Biology Innovator Grant. Work in progress.

    • C++
    • CUDA
    • OpenACC
    • HPC
    • Phylogenetics

Earlier projects

  • Automated Radiography Analysis Framework Using Deep Learning for Pneumonia and Covid-19 Identification

    Final year research & development project · Jun 2021 – Apr 2022

    • Python
    • TensorFlow
    • Computer Vision
    • CNN

  • Pump it Up: Data Mining the Water Table

    Semester 7 · Machine Learning · Aug – Sep 2021

    • Python
    • Machine Learning

  • Distributed Chat System

    Semester 7 · Distributed Systems · Sep – Oct 2021

    • Java
    • Netty

  • NCMS — WAKANDA

    SparkX · Individual assignment · Jun – Jul 2021

    • Java
    • Servlet
    • PostgreSQL
    • React

  • goSAFE

    ACM Student Branch, University of Moratuwa · May – Jun 2020

    • Android
    • Java
    • Firebase

  • Online Bus Booking System

    Semester 5 · Software Engineering project · Feb – Jun 2020

    • Node.js
    • Express
    • Firebase

  • SmartEdkit

    Code with WIE competition · Jul – Aug 2019

    • Flutter
    • Firebase

  • Healers

    Semester 3 project · Mar – Jun 2019

    • PHP
    • MySQL
    • HTML
    • CSS

Experience

  1. 2022 – 2024

    Software Engineer

    Sysco LABS Sri Lanka · May 2022 – January 2024

    • Learned new technologies on the job: Jenkins, Chef, AWS, Shell, Terraform.
    • Added new functionality and modifications to existing Jenkins pipelines.
    • Modified the existing Chef cookbook to support active-active highly available architecture.
    • Implemented a Lambda function to schedule starting and stopping AWS instances.
    • Modified existing Terraform code to provision new resources in AWS.
    • Designed the failover process for active-active highly available architecture.
    • Implemented new Jenkins pipelines supporting active-active highly available architecture.
  2. 2020 – 2021

    Software Engineer Intern

    CodeGen International (Pvt) Ltd · October 2020 – March 2021

    Database data security project

    • Implemented LDAP authentication for an in-house database security product that detects sensitive data, using Spring Security features.
    • Quickly learned the existing code and extended system capabilities at both the Angular front end and the REST-API backend.
    • Implemented support for multiple databases: MySQL and Oracle.

    Oracle APEX BI dashboard project

    • Set up Oracle Database and Oracle APEX environments and implemented an interactive BI dashboard for booking-related data analysis.

    NLP project

    • Annotated texts, updated gazetteer files and trained the in-house NLP engine to identify amenities from text.

Education

  1. Current

    Doctor of Philosophy

    School of Computing, ANU College of Systems & Society
    The Australian National University

  2. 2017 – 2022

    BSc Eng (Hons), Computer Science and Engineering

    University of Moratuwa · GPA 3.70 / 4.20
    First Class Honours · Dean's List for semesters 1, 6, 7 and 8

  3. 2021

    SparkX Professional Development Program

    SparkX Academy, Sri Lanka · Mar – Sep 2021

  4. 2014 – 2016

    GCE Advanced Level

    Devi Balika Vidyalaya, Colombo 08 · AAA · Island rank 158

Certifications

  1. 2025

    Fundamentals of Accelerated Computing with OpenACC

    NVIDIA · issued Feb 2025

  2. 2025

    Getting Started with Accelerated Computing in CUDA C/C++

    NVIDIA · issued Feb 2025

  3. 2025

    Introduction to Concurrent Programming with GPUs

    Coursera · issued Jan 2025

  4. 2021

    Microsoft Certified: Azure Fundamentals

    Microsoft · issued May 2021

  5. 2021

    Neural Networks and Deep Learning

    Coursera · issued Apr 2021

  6. 2020

    Developing Applications with Google Cloud Platform (specialisation)

    Coursera · issued Jul 2020 · four courses

  7. 2020

    Other Coursera and Udemy certificates

    Front-End Web UI Frameworks (Bootstrap 4) · AWS Fundamentals: Going Cloud-Native · API Design and Fundamentals of Google Cloud's Apigee API Platform · Web Application Security Testing with OWASP ZAP · Node JS API Development · React JS Frontend Web Development

Skills

Research areas

  • Phylogenetics
  • Bioinformatics
  • Machine Learning
  • Computer Vision
  • Parallel Programming
  • Distributed Systems

Parallel & HPC

  • CUDA
  • OpenACC
  • OpenMP
  • MPI

Languages

  • Python
  • C++
  • Java
  • JavaScript
  • PL/SQL
  • PHP

Cloud & DevOps

  • AWS
  • Google Cloud
  • Microsoft Azure
  • Terraform
  • Progress Chef
  • Jenkins

Web

  • Node.js
  • Spring Boot
  • React
  • Angular
  • HTML
  • CSS

Data & tooling

  • PostgreSQL
  • MySQL
  • MongoDB
  • Firebase
  • Git
  • Linux
  • Bash

Awards & activities

Grants

  1. 2026

    TALO Computational Biology Innovator Grant

    $10,000 over one year, for GPU-Accelerated Maximum Likelihood Phylogenetics in IQ-TREE 3: Enabling Large-Scale Evolutionary Analysis on Modern HPC Clusters · ANU TALO program

Awards & competitions

  • 2019

    Best Team Process

    Code with WIE — hackathon organised by IEEE WIE Sri Lanka Section

  • 2020

    Winners

    HackDown 2020 — coding competition organised by the IEEE WIE Student Affinity Branch of the University of Moratuwa

  • 2019 – 2021

    MoraXtreme

    12-hour programming competition organised by the University of Moratuwa

    • MoraXtreme 5.0 · rank 14
    • MoraXtreme 4.0 · rank 14
    • MoraXtreme 3.0 · rank 63
  • 2018 – 2020

    IEEEXtreme

    24-hour programming competition organised by IEEE

    • IEEEXtreme 14.0 · global rank 274
    • IEEEXtreme 13.0 · global rank 464
    • IEEEXtreme 12.0 · global rank 495

Service & outreach

  1. 2020 – 2021

    Webmaster

    IEEE Student Branch of the University of Moratuwa

  2. 2020 – 2021

    Webmaster

    Women in Engineering Symposium 2020 and 2021 — built the website for the International Women in Engineering Symposium using WordPress.

  3. 2020

    Creative Content Team volunteer

    IEEE SLSYW Congress 2020

Automated Radiography Analysis Framework Using Deep Learning for Pneumonia and Covid-19 Identification

Research & development project · June 2021 – April 2022

  • Python
  • TensorFlow
  • Machine Learning
  • Computer Vision
  • Convolutional Neural Network

An automated radiography analysis framework for pneumonia and Covid-19 identification, providing better performance in chest X-ray analysis for detecting lung infection conditions.

  • A systematic review presenting deep-learning-based pneumonia and coronavirus detection solutions, trends, datasets, guidance for a deep learning process, challenges and future research directions.
  • Chest X-ray classification with MobileNetV2, InceptionV3, Xception and ResNet50 models with added top layers.
  • An ensemble developed from the trained MobileNetV2, InceptionV3, Xception and ResNet50 models.
  • Experiments on chest X-ray classification with segmentation using a U-Net architecture.
  • A web application for the framework, combining segmentation and classification as a proof of concept.

Resulting publications

Four peer-reviewed outputs came out of this work — see the Research section.

Pump it Up: Data Mining the Water Table

Semester 7 · Machine Learning · Aug – Sep 2021

  • Python
  • Machine Learning

An individual project implementing a solution for the Pump it Up: Data Mining the Water Table competition on DrivenData. Covered pre-processing, encoding and feature engineering.

Models used

  • MLP Classifier
  • Random Forest
  • Decision Tree
  • XGBoost
  • Gradient Boosting Classifier

Distributed Chat System

Semester 7 · Distributed Systems · Sep – Oct 2021

  • Java
  • Netty

A distributed chat application. The goal was to build chat servers that clients can connect to and chat through — connecting with unique IDs, creating, joining and deleting chat rooms, and messaging other clients. Leader election uses the bully algorithm, failure detection uses heartbeats, and the Netty framework handles TCP connections.

My role

I developed the client requests for deleting a room, listing rooms, and moving between and joining rooms, and implemented part of the bully algorithm.

NCMS — WAKANDA

SparkX · Individual assignment · Jun – Jul 2021

  • Java
  • Java Servlet
  • PostgreSQL
  • React
  • React-Redux

A Covid management tool with authentication, bed reservation and Covid statistics review. The REST API was built with Java Servlets, the UI with React and React-Redux, and PostgreSQL as the database. Authentication uses JSON Web Tokens, password hashing uses jBCrypt, and logging uses Log4j.

goSAFE

ACM Student Branch, University of Moratuwa · May – Jun 2020

  • Native Android
  • Java
  • Firebase

An Android mobile application to track Covid-19 using Bluetooth technology.

My role

I handled the Firebase database and its relation functions. This was my first native Android application.

Online Bus Booking System

Semester 5 · Software Engineering project · Feb – Jun 2020

  • Node.js
  • Express
  • Firebase

A web-based system and two mobile applications with a REST API backend, letting passengers reserve bus seats online. A mobile application handles ticketing for the driver or conductor; the web application serves both bus owners and passengers. After payment, the passenger receives a receipt confirming the reservation and the conductor receives the reservation details.

My role

I developed the REST API using Node.js with Express and designed the Firebase database. This was the first REST API I built; I gained a lot of experience and used new tools such as Postman and Apache JMeter.

SmartEdkit

Code with WIE competition · Jul – Aug 2019

  • Flutter
  • Firebase

A web application for school administrators and a mobile application for school teachers, intended to improve the efficiency of the school system. It monitors attendance for both teachers and students, lets teachers take notes, and sends notifications reminding them of the next period.

My role

I developed the teachers' mobile application using Flutter. This was a minimum viable product built for the Code with WIE competition and my first mobile application.

Award

Team receiving the Best Team Process award at Code with WIE

Healers

Semester 3 project · Mar – Jun 2019

  • PHP
  • MySQL
  • HTML
  • CSS

A web application connecting students and mentors within the university. It eases communication between students and mentors — mentors can keep track of their students and assign tasks.

My role

I developed the authentication system, the messaging function and the mentor-selection function, both backend and frontend. This was the first web application I built. Frameworks were not permitted, so we built it on an MVC architecture — I learned the basics of web development from this project.