The People Behind the Project
Meet Our Team
Five engineers and analysts working together to bring intelligent waste classification to life — from raw datasets to a finished, deployable system.
Core Contributors
The Minds Behind REGAIA
Han Pyae Tun
He led the architectural design and implementation of the entire AI pipeline. He selected the CNN model architecture for image classification, defined end-to-end workflows, handled data preprocessing, and drove the model training process — ensuring the system's intelligence was built on a rigorous, well-engineered foundation.
Shin Myat Noe Zin
She curated the diverse dataset of waste images that underpins the model's performance. She oversaw data collection, maintained data quality standards, balanced class distribution across categories, and prepared the final dataset splits for training and testing — giving the model the varied, reliable input it needed to generalise well.
Lin Wai Yan
He built the user-facing layer of the system from the ground up. He designed and developed a fully responsive web interface that lets users upload waste images and receive instant classification results. His work prioritised smooth interaction, accessibility, and clean integration with the backend inference services.
Yu Nandar Hlaing
She focused on bringing all system components together into a cohesive, working whole. She conducted comprehensive system testing, tracked down and resolved bugs, and validated model outputs through the interface. She also contributed to hyperparameter tuning and evaluated key performance metrics such as accuracy and precision to ensure the system met quality targets.
Thant Htoo San
He was responsible for assessing the system's overall performance and capturing the project in writing. He produced detailed reports on model accuracy, usability findings, and known limitations. Alongside this, he authored the technical documentation, user guides, and final project reports — ensuring the work remains clear, reproducible, and accessible to future contributors.
Responsibilities at a Glance
How the Work Was Divided
Model & Data
Han Pyae Tun designed the neural network architecture and training strategy, while Shin Myat Noe Zin sourced and curated the dataset — together forming the core AI engine.
Interface & Integration
Lin Wai Yan built the responsive frontend and connected every layer of the stack, while Yu Nandar Haling handled the overall testing phase — making sure the model, backend, and UI all worked seamlessly as one.
Quality & Knowledge
Thant Htoo San evaluated system performance and translated the entire project into clear documentation — from technical specs to user guides — preserving the team's work for the future.