Solution Developer
Business systems on Dynamics 365 Business Central, and machine learning applications when the problem calls for one.
About
Day-to-day work centres on Microsoft Dynamics 365 Business Central: customising the interface, and connecting outside systems to it over REST APIs. That covers point of sale applications, desktop and console tools that post transactions and loyalty points, industrial scales feeding readings straight into Business Central, and a data warehouse running on SQL Server.
Before that came teaching a four-month PHP course, taking a cohort from the basics through to full-stack applications, and a traineeship building Flutter apps: the first work with computer vision and generative models.
The research side is not separate from the rest. The final year project, a custom convolutional neural network for lung cancer detection, reached 98.55% accuracy and was published at ICTAR. The same instinct, finding where a system falls short and fixing the cause rather than the symptom, carries over to business software.
Experience
Softserve Solution (Pvt) Ltd. · Promoted from Trainee Solution Developer
Customising and extending Dynamics 365 Business Central, and building the applications that talk to it: standalone desktop and console tools connected over REST APIs, a SQL Server data warehouse with automated backups, and every endpoint validated in Postman before it ships.
Lanka Corrugated Industries (Pvt) Ltd.
Documented the full corrugated manufacturing workflow, from sales order entry through to printing the advance order dispatch, and drew the flowchart that made the process legible to everyone who works in it. Also handled network, printer and hardware troubleshooting.
Academy of Giga Nerds
Designed and taught a four-month PHP for web development course, taking a cohort from first principles to full-stack applications. Covered MySQL, session management, secure coding and object-oriented design, built around projects that ended in working login systems and CRUD applications, with dedicated Git sessions throughout.
BevyLabs (Pvt) Ltd.
Built Flutter applications in an agile team, two of which reached Google Play. Wrote a computer vision app using YOLO and researched how far it could be pushed for image processing, and added generative features using Ollama and Google Gemini.
Enhancing Lung Cancer Diagnosis with a Custom-Designed Convolutional Neural NetworkICTAR 2024 · International Conference on Transformative Applied ResearchP. K. P. Sandamal and G. R. Perera · ICTAR, AI and Data Science Synergy track
A custom CNN for classifying lung CT scans, built to be smaller and faster to train than established pre-trained models. It reached 98.55% accuracy against 97.67% for ResNet50 and 66.15% for VGG16. Data came from Kaggle and Teaching Hospital Anuradhapura, balanced with SMOTE and augmentation, and evaluated by cross-validation on precision, recall and F1.
Read the paperUniversity of Plymouth, UK · Second Class Honours, Upper Division
Studied at NSBM Green University, Sri Lanka. Final year research: a custom convolutional neural network for lung cancer detection from CT scans, published at the ICTAR conference.
Research conference, AI and Data Science Synergy track
A paper on custom CNN design for lung cancer detection, presented and published, with the conference certificate awarded.
Hackathon completed, NSBM Green University
A working solution built and delivered to a brief under a hard time limit: a lesson in scoping and teamwork as much as in code.
Contact