Software engineer,
with eyes on the sky.
|
I'm Shameeha — a Software Engineering (Hons) graduate who builds web systems, AI-powered applications, and multilingual digital tools. My work spans production university platforms, medical AI, and research-driven software. I've led teams in both academic and professional settings, and I'm currently open to software engineering, AI/ML, research, and freelance opportunities. The stars on this site are not a gimmick — I've simply always found the night sky beautiful.
From bio-science to building intelligent systems.
My path into software was not a straight line — it was an orbit. I began in biological sciences, then pursued a Software Engineering (Hons) degree at the Open University of Sri Lanka while working in parallel as a data-entry operator for two years. That route shaped how I work: carefully, persistently, and with a strong respect for systems that need to be reliable for real people.
During my degree, I moved from learning software engineering to leading it in practice. I directed a 4-member intern team at the OUSL IT Division contributing production modules to the MyOUSL student portal, and I also led a 5-member final-year team building NeuroSight — an AI-powered brain tumour detection and patient-management platform that will be grow into published research.
I'm especially drawn to work that sits at the intersection of software engineering, AI/ML, and real-world usability: enterprise web platforms, multilingual digital tools, research-driven systems, healthcare technology, and applications that need to serve people clearly and reliably. I work in English, Sinhala, and Tamil, and some of the product decisions I'm proudest of are the ones that made a system more understandable across all three.
Outside of code, I write fiction and admire the night sky — as a happy amateur, not an astronomer. Longer term, I'm interested in how software and AI can support areas like medical decision support, environmental monitoring, and other high-impact problem spaces. For now, my focus is simple: build excellent systems, keep learning, and work with people who care about doing meaningful things well.
The journey, as a constellation
A sky map of what I work with.
Every cluster below is battle-tested — these are the tools behind shipped systems, not a wishlist.
Web & Frameworks
AI / Machine Learning
Languages
Databases & ORM
Blockchain & Storage
Tools & Practices
Systems that shipped.
NeuroSight — AI Brain Tumour Detection & Patient Management
Team leader of a 5-member team building an end-to-end clinical platform. I personally built the multilingual AI chatbot pipeline, the trilingual patient PWA, and the complete patient management website research based a real clinical workflow, while coordinating the deep-learning MRI classification work across the team.
What's under the hood
- WaveFusionNet — the team's EfficientNetV2 + DenseNet201 ensemble with dual-path Grad-CAM, so clinicians see why the model classified a tumour. My teammates led the model; I led the team and integration.
- Multilingual AI chatbot — fine-tuned DistilBERT with FAISS vector search, TF-IDF/SVC fallback, and emergency keyword detection, served as RESTful APIs via a FastAPI microservice.
- Tamper-proof records — doctor-confirmed results anchored on Ethereum (Sepolia) and IPFS via Pinata, with AES client-side encryption.
- Patient PWA — React + i18next, 70 education sections per language across English, Sinhala, and Tamil.
- Patient record management — a clinical web application handling patient records, doctor confirmations, and result workflows end-to-end.
MyOUSL Student Portal
Directed a 4-member Agile intern team shipping an enterprise-grade, production ASP.NET Core MVC (.NET 10) portal for thousands of OUSL students — delivering Exam Registration, My Program, Course Progress, and Results modules end-to-end with Clean Architecture, EF Core, and GitHub Actions CI/CD.
Intelligent Study Planner
AI-assisted study scheduler — predicts daily study hours with a rule-based engine plus optional Linear Regression, detects deadline clashes with automatic adjustment, and generates multi-day timetables with breaks and subject rotation.
Globetrotter
A self-help tourism Android application built in Java, designed to support travel planning and on-the-go exploration. It includes persisted trip data, map-based location support, compass-assisted navigation, and an offline-friendly mobile experience for travellers.
SMARTLIB
A group-built library management system and my first experience leading a software team. I coordinated the group and contributed to interface design for search and reporting workflows. Technically, it was an early project; in hindsight, its biggest value was teaching me how to guide a team, divide work, and keep a shared build moving — lessons that later shaped how I led NeuroSight and the MyOUSL internship team.
Pearl of the Indian Ocean
A responsive tourism showcase website built as a front-end web project, focused on presenting Sri Lankan destinations through clean layout, visual storytelling, and lightweight interaction — built with HTML, CSS, JavaScript, and Bootstrap.
Where the ocean meets the algorithm.
Detecting Underwater Macroplastic Pollution: A Review on AI-Based Computer Vision Methods and Their Relevance to Sri Lanka's Marine Ecosystem
Sri Lanka ranks among the world's top contributors to marine plastic waste, yet underwater macroplastic monitoring remains nearly unexplored here. This review synthesizes global AI and computer-vision approaches — YOLO, Faster R-CNN, U-Net, Mask R-CNN — against Sri Lanka's realities, and argues for lightweight, low-cost detection systems, a national annotated dataset, and AI-informed marine policy. It closes with a concrete roadmap toward a baseline AI monitoring network by 2027.
Research interests
I'm interested in research that combines AI, software systems, and real-world impact — particularly in areas such as:
- medical AI and clinical decision-support systems
- computer vision and explainable AI for sensitive domains
- multilingual AI systems and accessible digital health tools
- human-centred NLP for low-resource languages like Sinhala and Tamil
- AI-assisted education and student-support tools
- environmental monitoring, sustainability, and data-driven public-interest systems
An early-stage curiosity: I'm also interested in longer-horizon questions around space-weather monitoring and solar-storm prediction. At the moment, that interest is exploratory rather than established research work — but it reflects the kind of interdisciplinary problem space I'm drawn to: data, uncertainty, systems, and the challenge of building useful tools around them.
Let's build something worth looking up to.
Three kinds of messages make my day: software engineering, AI/ML, and data roles in Colombo or remote-friendly teams; research collaborations in medical AI, computer vision, or environmental monitoring; and freelance projects — if you need a website, web app, or AI-powered tool built, I'd love to hear what you're imagining.
I respond fastest by email, and I read everything personally.