Business Intelligence student at VŠE Prague, no-code / low-code developer, and data-extraction hobbyist. Most of what I build falls into three buckets: web scrapers, applied-AI services, and small desktop tooling.
- czech-tabloid-scraper — 2,400 headlines from 18 Czech news feeds in 14 seconds, stdlib + requests + rich
- chromadb-embedding-visualizer — 3D fly-through of ChromaDB vector stores. Next.js + Three.js frontend, FastAPI + UMAP + HDBSCAN backend
- email-verifier — async SMTP/DNS verification with catch-all detection and Czech-provider reputation logic
- vocative-generator — scripts sklonuj.cz to batch-decline Czech first names, with adaptive concurrency and resumable checkpoints
- czech-surname-restorator — Ray-distributed diacritic restoration for legacy Czech datasets (
Dvorak→Dvořák) - rm1000i-tray — tiny Windows menu-bar widget that reads live PSU power draw over USB HID
I'm open to projects in data extraction, workflow automation, and applied AI integrations (RAG, structured-output LLMs, agent-style pipelines).
Reach me at jan.alexandr.kopriva {@} gmail dot com.


