Hey, I'm
Developer based in Thessaloniki, Greece. I build full-stack web platforms, multi-cloud desktop tools, C++ systems libraries, and deep learning pipelines for medical imaging.
About Me
I'm Pavlos — a full-stack developer and ML engineer with a focus on building robust, well-engineered systems. My work spans enterprise web platforms, multi-cloud desktop tools, systems-level C++ libraries, and deep learning pipelines for medical imaging.
My thesis applied radiomic analysis and 3D convolutional neural networks — UNet, Dense UNet, and VGG16 transfer learning — to automated segmentation of lung cancer regions in CT scans.
When I'm not writing code, I'm exploring embedded systems, reading deep learning research, and contributing to open source.
Skills
Languages, frameworks, and tools I use across web, systems, and ML development.
Work
A selection of projects spanning web development, systems programming, and machine learning.
Full-stack project management platform with real-time WebSocket updates, Elasticsearch full-text search, bulk operations with optimistic concurrency, team roster management, and milestone tracking. Released v1.0.7 with 222 commits — includes published npm and PyPI packages.
Enterprise desktop application unifying Google Drive, OneDrive, and iCloud under a single interface. Features single sign-on via OAuth 2.0, real-time file synchronisation, and storage analytics — built with JavaFX 17 and Spring Boot 3.4.
Lightweight C++23 image processing and analysis library with minimal external dependencies. Supports Gaussian, Sobel, and median filters, geometric transforms, RGB↔HSV conversion, and thread-safe parallel processing. Formats: PNG, JPEG, BMP, PPM.
AI-powered platform for automated segmentation of CT scans, MRI, and X-rays using deep learning. Provides asynchronous analysis with progress tracking, an interactive Chart.js dashboard, and a JWT-secured REST API — fully containerised and CI/CD ready.
Diploma thesis on radiomic image analysis for automated segmentation of lung cancer regions in 3D CT scans. Implemented and benchmarked 3D CNN, UNet 3D, Dense UNet 3D, and VGG16 transfer learning architectures using TensorFlow/Keras.
Contact
Open to collaboration, job opportunities, or just a good conversation about tech.
Feel free to reach out through any channel below.