sergey osyka
Python automation for e-commerce & real estate
New to Contra
sergey is ready for their next project!
Production-ready Rolex tracker for Chrono24 ā built for portfolio, dealers, collectors and flip-market analysis. Demo mode included.
š„ What This Does
Track any reference ā edit TARGETS list or CLI --targets "Rolex Pepsi" "Rolex Batman"
Finds cheapest listing for each model (sorted by price on Chrono24)
SQLite price history ā perfect for trend charts, dashboards, analytics
New Lowest Price detection ā flags š„ NEW LOWEST! automatically vs. historical min
Daily CSV exports ā data/exports/rolex_prices_YYYY-MM-DD.csv + latest.csv
Anti-bot handling: rotating User-Agents, random delays (2-5s), persistent session
Self-healing parser: 3 fallback strategies + auto debug HTML dump if layout changes
Logging: daily rotating logs in /logs
Demo mode: --demo runs with realistic mock data ā no requests, perfect for portfolio review
Clean, PEP8, type-hinted, commented and easy to customize.
Stack: Python, Requests, BeautifulSoup + lxml, SQLite
1
23
From Airflow DAGs to a lightweight dashboard-only mode.
I developed an automated ETL pipeline for London weather data. The pipeline handles Extract (REST API polling), Transform (data cleaning & normalization), and Load (SQLite with timestamps). The dashboard shows Current Temp 23.3Ā°C, PM2.5 4.30 Āµg/mĀ³, historical trends, and raw data table.
Key decisions: switched from full Airflow setup to a lightweight Streamlit app for faster deployment and lower resource usage, while keeping the code clean and reproducible.
Repo structure: dags / src / dashboard / data / scripts - ready to be used as a template for any weather/IoT data project.
1
24
Automated Financial ETL: Replaces manual spreadsheet downloads with a validated, logged pipeline that delivers clean currency rates at 7 AM daily.
1
0
36
gmaps-leads-scraper-pro
1
0