PhD candidate at the University of Melbourne's CLOUDS Lab, building distributed systems for sustainable cloud datacenters. My work spans the full stack: cloud orchestration, Linux operating systems, and CPU-level power hardware. Former enterprise middleware engineer at WSO2. Looking to join a team building distributed infrastructure at scale.
Education
-
Ph.D. in Computer Science
University of Melbourne, Melbourne, Australia
2021–Present
Distributed systems for sustainable cloud computing. Melbourne Graduate Research Scholarship recipient. Advised by Prof. Rajkumar Buyya, Dr. Shashikant Ilager (TU Wien), and Dr. Maria A. Rodriguez.
-
B.Sc. Engineering (Hons.), Electronic & Telecommunication
University of Moratuwa, Sri Lanka
2014–2018
First Class Honours. Built a custom-ISA processor on an FPGA in Verilog and a 3D-convolutional video action-recognition system. Foundations in computer architecture, networks, and machine vision.
Experience
-
Graduate Researcher (PhD)
Building distributed systems for sustainable cloud datacenters, spanning the full stack from cloud orchestration down to host operating systems and CPU hardware — the middleware-to-node perspective.
- Designed and implemented carbon-aware scheduling that spans OpenStack orchestration down to per-core CPU power management on Linux hosts.
- Built a Go service over Intel's Power Optimization Library to drive per-core frequency/power features, bridging hardware power control with the cloud control plane.
- Evaluated systems both as working prototypes on commodity hardware and via large-scale simulation; published peer-reviewed results (IEEE CLOUD, SPE).
- Served on the artifact evaluation committee for USENIX ATC '23; teaching assistant leading tutorials, marking, and assignment design.
-
Senior Software Engineer
R&D on the open-source WSO2 Identity Server — enterprise-scale identity and access management middleware.
- Led subteams within the product team and represented the company as the senior CIAM engineer on one of its largest integration projects — a signature US hospitality brand.
- Ran production systems for enterprise customers: on-site deployments, performance and incident support, and customer training.
- Two-time recipient of WSO2's Outstanding Contribution award.
-
Software Engineer
Engineer on the R&D team for the open-source WSO2 Identity Server.
- Contributed to a large open-source Java codebase serving enterprise IAM deployments.
- Gained hands-on experience running enterprise production systems, customer support, and on-site deployments.
-
Software Engineering Intern
Undergraduate internship building a real-time computer-vision pipeline.
- Built a C++ system that processes a live satellite-TV stream and runs a deep image-classification model in real time to embed content-aware advertisements.
Research
- Aging-aware CPU Core Management for Embodied Carbon Amortization in Cloud LLM Inference Pre-print · 2025
- A Framework for Carbon-aware Real-Time Workload Management in Clouds using Renewables-driven Cores Pre-print · 2024
- DEMOTS: A Decentralized Task Scheduling Algorithm for Micro-Clouds with Dynamic Power-Budgets IEEE 16th International Conference on Cloud Computing (CLOUD), Chicago, IL, USA · 2023
- CloudSim Express: A novel framework for rapid low-code simulation of cloud computing environments Software: Practice and Experience (Vol. 54, Issue 3) · 2023
Projects
-
openstack-gc
OpenStack extension for carbon-aware computing: harnesses intermittent renewable energy for real-time VMs on commodity datacenter hardware.
-
core-power-mgt
A Go service wrapping Intel's Power Optimization Library to manage per-core CPU power features — wiring hardware power control into the cloud control plane.
-
orion
An API-driven, Java distributed computing framework: REST task submission, a worker pool executor, and a monitoring app. Featured in Towards Data Science.
-
cloudsim-express
A low-code framework for rapid simulation of cloud computing environments. Published in Software: Practice and Experience.
-
deep-video-activity-recognition
Real-time activity detection in a video feed using a deep 3D-convolutional model (Python).