cv

Software Engineer

Stamplus Rewards - Richmond, BC, Canada

Led sprint planning, product documentation, database design, and API design of a mobile rewards platform for merchants in Metro Vancouver region.

• Utilized AWS cloud computing services (EC2 and S3) to deploy and operate Python-based (Django) backend with RESTful API to facilitate CRUD operations in PostgreSQL DB.
• Slashed rewards collection time during checkout by 25% by gathering stakeholder feedback and designing auto-expiring QR codes.
• Improved user engagement by 20% by incorporating third-party features, including social media platform integration using Facebook and Google APIs and push notifications using Firebase.
• Contributed to frontend mobile development for reward collection and redemption using TypeScript (Ionic Framework with React).

Junior Software Developer

Archiact Interactive - Vancouver, BC, Canada

Developed and maintained VR/AR games and applications for Oculus Rift, PSVR, Samsung Gear VR, Android, and iOS.

• Collaborated cross-functionally with UI/UX designers to build custom, internal development tools using C# (Unity) for Marvel: Dimension of Heroes; improved UI/UX-related development speed by 50%.
• Converted existing codebase of non-VR/non-AR games into VR/AR compatible versions published on various platforms, titles include Waddle Home, Darknet, and Slots in De Nile.

An Adaptive Heuristic-Based Framework to Enhance JITServer Technology

Contributed to IBM-funded, open-source research projects focused on improving enterprise Java application performance in cloud computing environments using JITServer remote compilation technology.

• Spearheaded experiments to improve microservice web application startup time by 10% via reducing Java container image size (by up to 50%) automatically using Python scripts.
• Analyzed benchmark applications (Spring and OpenLiberty) using Bash, C++, and Python to identify up to 18% of JIT compilations can be further optimized to improve Java application performance in Eclipse OpenJ9 JVM.
• Led development of a visualization tool to aid understanding of Java compilation and optimization behaviour for developers; used JavaScript (vis.js), HTML, CSS (Bootstrap).
• Published international conference papers; received the best paper award at CASCON 2024 as main author.

Developing a Graduate Course using Eclipse Open J9 and OpenShift

Developed hands-on labs and documentation for an award-winning course on self-adaptive software, with practical exercises using OpenLiberty, MongoDB, Docker, Kubernetes, Prometheus, GCP, and IBM Cloud Observability.

Student Volunteer

CASCON 2020

Student Volunteer

CASCON 2019

PhD

University of Waterloo

Computer Engineering

MEng

University of Waterloo

Computer Engineering

Bachelor of Applied Science

University of British Columbia

Computer Engineering

Best Paper Award

CASCON 2024

Received the best paper award as the main author.

Postgraduate Scholarship - Doctoral Program

Natural Sciences and Engineering Research Council of Canada

Received federal scholarship based on submitted research proposal.

Beyond the Classroom: Bridging the Gap Between Academia and Industry with a Hands-on Learning Approach

Xu, M., Liu, R., Stoodley, M., Tahvildari, L. (CSEE&T)

Presented experience teaching a course on self-adaptive software systems that integrates theoretical knowledge and hands-on learning with industry-relevant technologies.

Using Semeru Cloud Compiler to Enhance Cloud-Native Java Application Performance

Liu, R., Shinde, S., Tahvildari, L., Stoodley, M., Sundaresan, V., Pirvu, M. (CASCON)

Proposed methods to extend the Semeru Cloud Compiler (SCC) to reduce the CPU and memory overhead of JIT compilation in containerized Java applications, and presented empirical evaluation results.

FlaKat: A Machine Learning-Based Categorization Framework for Flaky Tests

Lin, S., Liu, R., Tahvildari, L. (Arxiv)

Developed ML-based pipelines for fast and accurate flaky testing categorization of Java unit tests using Python and scikit-learn, which can be integrated into CI/CD workflows.

Using POMDP-based Approach to Address Uncertainty-Aware Adaptation for Self-Protecting Software

Liu, R., Tahvildari, L. (Arxiv)

Modeled state uncertainty and model parameter uncertainty within a data-driven Moving Target Defense deployment process using Reinforcement Learning and Bayesian Machine Learning techniques.

AHA: Adaptive Hadoop in Ad-hoc Cloud Environments

Liu, R., Lin, S., Tahvildari, L. (ACSOS)

Designed data-driven Resource-aware Task Scheduler (using Java) for running distributed computing within ad-hoc cloud environments; improved performance by up to 20.2%.