Hi, 

I'm Michael Selvesakis

Full Stack Developer, System Administrator, and Cybersecurity Enthusiast

About Me

I'm a software engineer based in Kavála, Greece. I graduated in Informatics from the Democritus University of Thrace in 2026, with a thesis benchmarking LLM-based recommenders against classical algorithms on accuracy, energy and cost.

I work as a Full Stack Developer at Classter on a multi-tenant school and university management platform — C#/.NET on the product side, Python and TypeScript on the AI side — and handle web development and system administration for IQ3SOLAR in Cyprus.

I have also published twice, on educational technology and on sustainable machine learning.

Experienced With:

2026

Full Stack Developer

Work

Classter

Oct 2025 - Present

C#/.NET feature work on the Consumer API and admin Portal, over SQL Server on Azure. I also own the AI and documentation platform — an in-product retrieval-augmented assistant, an MCP server, and the docs-as-code system feeding both.

Back End Developer, Intern

Work

Classter

May 2025 - Aug 2025

Back-end internship on the same platform, working in C#/.NET against the SQL Server estate. Converted to a permanent full-stack role in October 2025.

Founder Member

Volunteer

OpenSource DUTH

Jan 2025 - Sept 2025

Founding member of the OpenSource DUTH community. Organized projects promoting open-source software and contribution, and planned workshops and events.

Full Stack Web Developer & System Administrator

Work

IQ3SOLAR - Limassol, Cyprus

Mar 2024 - Present

Building an interactive web application that pairs PV algorithms with production data to forecast future energy output and surface faults in solar installations. Also responsible for system administration, servers and networking.

System Administrator

Volunteer

GridLab - DUTH

Oct 2023 - Sept 2025

Volunteered as a System Administrator on university infrastructure, managing servers and deploying applications.

B.Sc. in Informatics

Education

Democritus University of Thrace

2021 - 2026

Thesis: "Comparative Evaluation of Traditional Recommender Systems and LLM-based Recommenders" (10/10). Benchmarked 19 recommender systems across accuracy, ranking quality, energy footprint and compute cost.

Projects

This is a collection of some of the projects I have worked on. Click on each card to find out more about each project.

Publications & Credentials

Peer-reviewed work, and the certifications behind the infrastructure half of what I do.

October 2025

Beyond Accuracy: Benchmarking Machine Learning Models for Efficient and Sustainable SaaS Decision Support

Future Internet 17(10), 467

A peer-reviewed journal paper arguing that model selection for SaaS decision support should weigh efficiency and sustainability alongside predictive accuracy.

Mavridou, E., Vrochidou, E., Selvesakis, M., & Papakostas, G. A.

July 2024
First author

SQLatch: Enhancing SQL Learning through Interactive Block-Based Programming

ICETOL 2024, Eskişehir, Turkey (pp. 46–51)

A block-based programming environment that lowers the barrier to learning SQL, presented at the 4th International Conference on Educational Technology and Online Learning.

Selvesakis, M., Harizanis, G., & Kazanidis, I.

2022 – 2025

Cisco Networking & Security track

Cisco

Six credentials: Introduction to Cybersecurity, Networking Basics, Networking Devices and Initial Configuration, Network Addressing and Basic Troubleshooting, Network Support and Security, Network Defense — completing the Network Technician career path.

2025

Microsoft Azure AI Essentials Professional Certificate

Microsoft & LinkedIn

2024

Univators: Skilling Future Digital Innovators

Code.Hub

2026

Pedagogical & Teaching Competence

Democritus University of Thrace

Let’s Talk!

Feel free to reach out! I'm always open to discussing new projects and opportunities.

Send me a message