Dr. Michail Stamatis

Dr. Michail Stamatis

Data Engineer

PhD in Atmospheric Physics

About Me

I'm a Data Engineer at NIKI Digital Engineering (partner: Audi, BMW) with a PhD in Atmospheric Sciences. My early research on climate and radiation datasets sparked my interest in building systems that turn complex data into something practical and usable. Today, I design scalable ETL pipelines and applications using Python, PySpark, SQL, and AWS — cutting report latency and automating workflows for the German automotive industry. I also build full-stack web applications as a freelancer and teach Python to aspiring data scientists. Outside of work, I teach physics and math to high school students, and enjoy stargazing, cycling, gardening, and exploring Greek history.

Technical Skills

Python Primary language — used daily for ETL pipelines, data engineering, automation, and full-stack data apps. PySpark Processing multi-TB datasets in distributed AWS environments; core tool for large-scale ETL at NIKI Digital Engineering. SQL Complex queries for pipeline transformations, analytics, and cloud database management across PostgreSQL and AWS Athena. AWS Building and orchestrating cloud data infrastructure — S3, Glue, Athena, Lambda — for production automotive pipelines. Airflow Orchestrating multi-step data workflows and scheduling pipeline jobs in production environments. Docker / Git Containerised, reproducible data engineering environments and version-controlled pipelines ready for team collaboration. Linux / Bash Daily shell scripting, job scheduling, and system administration — comfortable in HPC and cloud environments alike. FastAPI Building lightweight data APIs and backend services — used in freelance projects to serve processed data to front-end clients. Pandas / Dask Data wrangling and transformation at both small and large scale — from exploratory analysis to production pipeline preprocessing. Java Familiar with Java in data engineering contexts — reading and adapting JVM-based pipeline code and working alongside Java-heavy stacks. JS / HTML / CSS Front-end layer for freelance data apps — building interactive interfaces that make data products accessible to end users. Scikit-learn / PyTorch ML prototyping and statistical modelling — applied to radiative transfer outputs during PhD research and pipeline anomaly detection. Fortran Maintaining and running high-performance radiative transfer models in scientific computing environments. Power BI Building dashboards that translate complex pipeline outputs and KPIs into readable visuals for non-technical stakeholders.

Domain Expertise

Data Engineering Designing end-to-end ETL pipelines from raw satellite/automotive data to production-ready datasets. Data Analytics Turning messy, large-scale data into clear business and scientific insights using statistics and visualisation. Software Development Building reliable, maintainable tools — from CLI utilities to full-stack scientific web applications. Atmospheric Sciences PhD-level expertise in climate variability, surface solar radiation, and global dimming & brightening. Quality Control Systematic validation of climate datasets against 1,000+ ground-truth stations worldwide. Machine Learning Applied ML for scientific emulation (Random Forests, PINNs) and exploratory data analysis.

Education

PhD

Sep 2020 – Sep 2025

University of Ioannina

PhD in Atmospheric Sciences and Environment — Grade: 10/10
  • Thesis: Detailed assessment of global dimming and brightening under all-sky and clear-sky conditions using modern tools and long-term climate data
  • Interdecadal changes of Surface Solar Radiation
  • Radiation Transfer Calculations
  • Remote Sensing Data Analysis
  • Quality Control and Evaluation of climate datasets
  • Software Development, Machine Learning
  • CLARISC database, 4 first-author peer-reviewed papers, 9 conferences

EUMETSAT Autumn School

Oct 2023

Athens, Greece

Remote Sensing Data Applications

Master's Degree

Sep 2016 – Jun 2019

Aristotle University of Thessaloniki

MSc Environmental Physics — Grade: 8.64/10
  • Thesis: African Dust Exports to Greece
  • Physics and Chemistry of the Atmosphere
  • Programming languages: Fortran, IDL, Matlab, R

Bachelor's Degree

Sep 2012 – Sep 2016

University of Ioannina

Physics Degree — Grade: 7.27/10, Top 2%
  • Classic and Modern Physics, Meteorology
  • Statistics, Calculus, Chemistry
  • Programming Languages: C/C++, Fortran

Experience

Data Engineer

Jun 2024 – Present · Full Time

NIKI Digital Engineering (External partner: Audi, BMW) · Remote

  • Built AWS ETL pipelines (PySpark + SQL), cutting report latency by ~30%
  • Automated ECU test workflows in Python and EXAM, reducing manual effort by ~40%
  • Prototyped Physics-Informed Neural Networks (PINNs) for scientific simulation applications
  • Contributed to proposal writing for EU-funded and national engineering projects
  • Data Analyst: Power BI, Python, Pandas, Excel, SQL

Scientific Data Engineer & Analyst — PhD Researcher

Oct 2020 – Sep 2025 · Full Time

Laboratory of Meteorology and Climatology, University of Ioannina

  • Architected CLARISC — cloud database integrating EUMETSAT CLARA & NASA ISCCP-H satellite datasets
  • Engineered multi-TB pipelines for satellite and reanalysis products (Python, Fortran, Dask, SQL)
  • Applied ML & statistics to filter and analyse Radiative Transfer Model outputs
  • Published 4 peer-reviewed papers; presented at international conferences
  • Built an interactive web application (EarthSense) for reproducible access to PhD results

Python Instructor — MSc Atmospheric Sciences

Mar 2022 – Present · Part Time

Laboratory of Meteorology and Climatology, University of Ioannina

  • Taught applied Python: Pandas, NumPy, xarray, time series, netCDF
  • Improved student coding proficiency by ~45% across 3 cohorts

StamazingApps — Freelancer

2024 – Present · Part Time

  • Scientific Web Apps & Full-Stack Applications

IT Research Services

2022 – 2023 · Full Time

Dioni: Computing Infrastructure for Big-Data Processing, University of Ioannina

  • Led high-performance computations for processing and analysing up to 20 TB of data
  • Oversaw meteorological software installation and execution for optimal performance

High School Science Tutor

2016 – Present · Part Time · Freelancer

  • Teaching Physics, Maths and Chemistry to high school students

Projects

ERMES

ERMES

ERA5 and CAMS Meteorology Explorer System.

EarthSense

EarthSense

My PhD made interactive through this web app!

Aether

Aether

Fast NetCDF Explorer.

NATEX

NATEX

.nat files explorer

Cyclops

Cyclops

Live weather conditions via OpenWeatherMap.

Phoebe

Phoebe

Relationship ML Explorer.

Hyperion

Hyperion

A Random Forest SSR Emulator trained with EarthSense data (Stamatis et al., 2025).

ChristmasApp

ChristmasApp

Visualize Timeseries: Christmas Edition

Apollo

Apollo

📈 Forecasts for Stocks, Crypto & Forex

Research

Which are the main drivers of Global Dimming and Brightening?

Stamatis, M.; Hatzianastassiou, N.; Korras-Carraca, M.-B.; Matsoukas, C.; Wild, M.; Vardavas, I.

Atmospheric Research, Volume 322, 2025, 108140. https://doi.org/10.1016/j.atmosres.2025.108140

The Global Dimming and Brightening (GDB) phenomenon plays an important role in the Earth's climate, with clouds and aerosols being the major drivers. This study investigates GDB causes by quantifying the contributions of changes in clouds, aerosols, water vapor and ozone to the surface solar radiation (SSR) changes during 1984–2018. To this aim, radiative transfer calculations were performed by the FORTH-RTM on a monthly basis and 0.5°x0.625° spatial resolution using modern and improved datasets for clouds and aerosols. Validation against high-quality ground measurements confirmed RTM's reliability. Results show a global mean brightening of 0.88 Wm−2decade−1 from 1984 to 2018, stronger over land (2.57 Wm−2decade−1) than oceans (0.19 Wm−2decade−1). Globally, changes in clouds were the main GDB drivers. However, the contribution of aerosol optical depth (AOD) changes was remarkable over specific land areas with strong anthropogenic activity, such as Europe, India and East China.

How strong are the links between global warming and surface solar radiation changes?

Stamatis, M.; Hatzianastassiou, N.; Korras-Carraca, M.-B.; Matsoukas, C.; Wild, M.; Vardavas, I.

Climatic Change 177, 156 (2024). https://doi.org/10.1007/s10584-024-03810-6

The aim of this study is to investigate the possible relationship between the recent global warming and the interdecadal changes in incoming surface solar radiation (SSR), known as global dimming and brightening (GDB). The analysis is done on a monthly and annual basis on a global scale for the 35-year period 1984–2018 using surface temperature data from ERA5 reanalysis and SSR fluxes from the FORTH radiative transfer model. Our analysis shows that SSR fluctuations affect global warming rates. During the dimming phase in the 2000s, warming rates slowed down over Europe and East Asia, while during brightening phases warming rates were reinforced. Although GDB is not the primary driver of recent global warming, it can affect warming rates, partly counterbalancing or accelerating the dominant greenhouse warming.

An Assessment of Global Dimming and Brightening during 1984–2018 Using the FORTH Radiative Transfer Model and ISCCP Satellite and MERRA-2 Reanalysis Data.

Stamatis, M.; Hatzianastassiou, N.; Korras-Carraca, M.-B.; Matsoukas, C.; Wild, M.; Vardavas, I.

Atmosphere 2023, 14, 1258. https://doi.org/10.3390/atmos14081258

In this study, an assessment of the FORTH radiative transfer model (RTM) surface solar radiation (SSR) as well as its interdecadal changes is performed during 1984–2018. A thorough evaluation against high-quality reference surface measurements from 1193 GEBA and 66 BSRN stations is conducted. For the first time, the FORTH-RTM delta(SSR) was evaluated over an extended period of 35 years at 0.5°×0.625° spatial resolution. The RTM deseasonalized SSR anomalies correlate satisfactorily with GEBA (R=0.72) and BSRN (R=0.80). Results indicate a considerable and statistically significant increase in SSR (Brightening) over Europe, Mexico, Brazil, Argentina, Central and NW African areas from the early 1980s to the late 2010s.

Interdecadal Changes of the MERRA-2 Incoming Surface Solar Radiation (SSR) and Evaluation against GEBA & BSRN Stations.

Stamatis, M.; Hatzianastassiou, N.; Korras-Carraca, M.-B.; Matsoukas, C.; Wild, M.; Vardavas, I.

Appl. Sci. 2022, 12, 10176. https://doi.org/10.3390/app121910176

This study assesses and evaluates the 40-year (1980–2019) MERRA-2 surface solar radiation (SSR) as well as its interdecadal changes against high-quality reference measurements from 1397 GEBA and 73 BSRN stations. The study is innovative in that MERRA-2 delta(SSR) has never been evaluated before, and the SSR fluxes have never been evaluated at this global scale over 40 years. The MERRA-2 deseasonalized SSR anomalies correlate well with GEBA (R=0.61) and BSRN (R=0.62), evaluated at temporal scales spanning from decadal sub-periods to 40 years.

Blog

GRECHO

GRECHO

Visit my Blogpost

Stories from Byzantium and Early Modern Greece

Read More
Bokeh

Bokeh Data Viz Showcases

Visit my Posts

Innovative and advanced data visualisation showcases with Bokeh.

Read More

Get In Touch

I'm always open to discussing new projects, opportunities, or collaborations.

mixstam1453@gmail.com
⬇ Download CV