Zahary Kwidzynski
Aberdeen,Aberdeen
Summary
Postgraduate full-stack developer with experience building robust Android and web-based systems. Skilled in Kotlin, Java, Spring Boot, SQL, and front-end technology including React Native. Delivered end-to-end applications including an NHS-backed teledermatology platform with backend integration and AI-assisted features. Experience working in a team environment in university and internship. Familiar with CI/CD, Docker, Git workflows, and mobile-first design. Also experienced with AI and computer vision for applied use cases and delivery of certificate-based projects.
Overview
2026
2026
years of professional experience
Work history
Master's Thesis: NHS Electronic Health Record Skin Care System
University of Aberdeen
01.2025 - 07.2025
- Developed full-stack mobile teledermatology application using Android Studio to integrate AI-powered solutions for skin condition analysis.
- Worked on top of existing code-base to implement custom YOLOv12 instance segmentation models utilising custom dermatology dataset.
- Developed custom IQA metric system for calculating skin lesion quality using contrast, brightness, sharpness, and blurriness.
- Currently exploring model quantisation techniques to optimise AI performance for mobile deployment.
Lightweight Fish Instance Segmentation Research
University of Aberdeen
01.2024 - 07.2024
- Utilised Jetson Orin Nano edge device to develop lightweight fish instance segmentation model that outperformed existing research based state-of-the-art models in MMDetection on a closed-source fish dataset, scoring 79.5% in mean average precision.
- Utilised contrastive learning and notable neck alterations with CARAFE up-scaling technology to produce performance benefits.
- Incorporated elements of research-based components and adjusted model's head, neck and backbone.
- Comparative data analysis between performance metrics and memory usage for usage on ship.
Automated 3D Plant Segmentation Pipeline
University of Lincoln
- Developed and optimised machine learning data pipeline to segment 3D generated plants using PointNet.
- Integrated custom datasets into framework and automated processes.
- Labelled 3D plant images using Segments.ai.
- Modified existing PointNet repository to match custom dataset structure.
- Identified and resolved training-time issues such as infinite looping at batch-wise level and hard coded variables.
- Integrated annotation and texture point clouds to organise line-by-line data into appropriate repositories.
Education
Master of Engineering (MEng) - Computing Science/Computer Science
University of Aberdeen
AberdeenSkills
- Git/GitHub
- Linux (Ubuntu)
- Docker
- Android Studio
- Maven
- Python
- Java
- C
- C#
- Ruby
- SQL
- Kotlin
- JavaScript
- Numpy
- OpenCV
- Scikit-learn
- Pandas
- Matplotlib
- React-Native
- Detail focused
- Problem-solving
- Agile
- SCRUM familiarity
- DevOps principles familiarity
- API integration and development
- User interface design
- Backend development techniques
- AWS experience
Timeline
Master's Thesis: NHS Electronic Health Record Skin Care System
University of Aberdeen
01.2025 - 07.2025
Lightweight Fish Instance Segmentation Research
University of Aberdeen
01.2024 - 07.2024
Automated 3D Plant Segmentation Pipeline
University of Lincoln
Master of Engineering (MEng) - Computing Science/Computer Science
University of Aberdeen
Similar Profiles
Thomas DaniyaThomas Daniya
Doctoral Researcher at University of AberdeenDoctoral Researcher at University of Aberdeen
Shaokai TuShaokai Tu
Teaching and Operations Assistant at University of AberdeenTeaching and Operations Assistant at University of Aberdeen
Matthew G. ThomsonMatthew G. Thomson
PGDE Student Placements at University of AberdeenPGDE Student Placements at University of Aberdeen
Ian NivenIan Niven
Offshore Electrical Surveyor at DNVOffshore Electrical Surveyor at DNV
OLUMIDE Ebenezer AOLUMIDE Ebenezer A
Desktop Support Engineer at Greensafe Inteligent (Vanquish Bank)Desktop Support Engineer at Greensafe Inteligent (Vanquish Bank)