kkartas.gr personal homepage

Kyriakos Kartas

Software Engineer

Hi, I’m Kyriakos. I work on backends, databases and integrations.

Most of my work involves Python, SQL Server and REST APIs: moving data between applications, connecting business systems and building the services behind them.

I also work on research software: testing ETL pipelines, preparing meteorological data and evaluating machine-learning models. My publications and MSc research are listed below.

You can find my code on GitHub or read my CV for my professional background.

location
Thessaloniki, Greece
focus
backend / data / integration

Projects

all repositories ↗

Personal projects and integration tools. Source code linked below.

APILinker

Python

A REST-to-REST integration framework with field mapping, transformations, pagination, scheduling and retries. Supports multi-source joins and optional message-queue connectors.

API integration / automation / data mapping

documentation · PyPI

ETLForge

Python

ETL toolkit with schema-based data generation and validation. Built for working with structured data.

data engineering / ETL / validation

PyPI · JOSS paper

MetDataPy

Python

Prepares weather-station time series for analysis and machine learning: ingestion, unit normalization, quality-control flags, derived weather metrics and chronological data splits.

meteorology / time series / reproducible preprocessing

documentation · PyPI

Investment-Analysis-Tool

Python

A Flask application for exploring stock market data and technical indicators.

Flask / market data / REST API

Galaxy2Woocommerce

Python

Connects Galaxy ERP with WooCommerce through REST APIs.

ERP / e-commerce / integration

Galaxy2Opencart

Python

Connects Galaxy ERP with OpenCart. An integration between back-office software and an online store.

ERP / e-commerce / integration

Journal publications

  1. Towards Trustworthy Data Pipelines: A Maturity Model for Justified Reliance

    Kyriakos Kartas. International Journal of Computer Applications, 187(117), 1–12. June 2026.

    doi: 10.5120/ijcaa08a422f79a1

  2. ETLForge: A unified framework for synthetic test-data generation and ETL validation

    Kyriakos Kartas. Journal of Open Source Software, 11(119), 9326. March 2026.

    doi: 10.21105/joss.09326

Preprint

MSc & thesis

MSc Computer Science

University of East London · 2025–2026

MSc thesis · May 2026

Τοπική βραχυπρόθεσμη πρόγνωση καιρού με machine learning και deep learning

English title translation: Local short-term weather forecasting with machine learning and deep learning.

Air-temperature forecasting from a single Weathercloud station using only historical observations. The study compares 11 baseline, machine-learning and deep-learning models across six forecast horizons, from 10 minutes to 24 hours.

The Python pipeline uses MetDataPy for preprocessing, chronological data splits and training-only scaler fitting. Random Forest performed best at 1 and 3 hours; Gradient Boosting led at 6, 12 and 24 hours. The evaluated deep-learning models did not outperform the strongest tree-based models in this experiment.

Download thesis [PDF, 2.0 MB] · code & experiments ↗

82 pages · Greek, with an English abstract

Technical focus

Backend
Python services and REST APIs.
Databases
SQL Server, queries and data processing.
Integration
ERP systems, e-commerce platforms and the data exchanged between them.
Interests
System architecture, DevOps and cloud technologies.

Contact

For work or a conversation about software, reach me on LinkedIn.

For code, head to github.com/kkartas.

Download CV [PDF]