I build end-to-end scientific and AI systems — from field measurements and mathematical models to APIs, web platforms and cloud deployment. Founder of GeoHydroAI, hydrologist and PhD researcher, Python instructor at Beetroot Academy. More about me →
From physical evidence to verified knowledge and practical tools.
- GeoHydroAI — bilingual flood platform for Ukrainian river basins: public flood map, HAND screening, GR4J/GR4H runoff runs, evidence-cited knowledge bot.
- Model Runtime — self-hosted control layer for open-weight LLMs: on-demand vLLM / llama.cpp, one OpenAI-compatible gateway, 338 tests.
- GeoHydroAI Knowledge — research chat on geohydroai.org: RAG over ~7,000 papers and their tables plus a knowledge graph; every sentence cited, unsupported ones removed.
- DEM & Floods — where global elevation models fail and how that changes flood maps.
- Teaching — 52-lesson Python course with PR-based code review · course book.
Engineering · Research · Teaching
Validation of global elevation models using ICESat-2 LiDAR data for floodplain modeling in the Ukrainian Carpathians
Geophysical Journal · 2026 · doi:10.24028/gj.v48i2.346831
A High-Resolution Hydrological Dataset for Ukrainian River Basins With an Interactive Web Interface
Geoscience Data Journal · 2025 · doi:10.1002/gdj3.70027
All publications: Google Scholar · ORCID · Research overview
PhD research at the Ukrainian Hydrometeorological Institute, supported by the German Federal Environmental Foundation (DBU) with the Bavarian Environment Agency (LfU) as partner.
Python · R · JavaScript · TypeScript · React · FastAPI · Django · PostgreSQL/PostGIS · Docker · AI/ML · GeoAI
geohydroai.org · About · Google Scholar · LinkedIn · ORCID