Serhii D.
Offer Serhii work on your next project.
Private message
Offer work
Kropivnitskiy,
Ukraine
17 days 17 hours back
available for hire
on the service 2 months 26 days
Rating
Successful projects
Average rating
Skills and abilities
Portfolio
-
18 USD Rozetka scraping
Data Parsing
A reliable web scraper, ready for use in a production environment, designed to extract data about laptops from the largest Ukrainian e-commerce platform (Rozetka.ua), with automatic cloud synchronization and advanced bot anti-detection bypassing.
Key technical features:
… Cloudflare and bot bypassing: Based on Playwright and playwright-stealth to simulate real human behavior, allowing successful bypassing of strict anti-bot systems and Turnstile CAPTCHAs.
Fault-tolerant architecture: Integrated with an SQLite database to track scraping progress page by page. If the script is interrupted or fails, it automatically resumes from the last page.
Automatic cloud synchronization: Direct integration with the Google Sheets API (gspread) for real-time data dumping and formatting.
Smart filtering: Pre-configured backend filters (1 TB SSD, price < 25,000 UAH, specific leading brands).
Technology stack: Python 3.10+, Playwright, Playwright-Stealth, SQLite, Google Sheets API.
This project demonstrates clear code architecture, proper API integration, database state tracking, and professional web automation skills.
-
33 USD eBay scraping
Data Parsing
Asynchronous scraper for laptops on eBay, built on Python, Playwright, SQLite, and Google Sheets API.
This project automatically collects data about laptops from eBay, using dynamic page rendering through the proxy API Scrape.do. The scraper extracts detailed product information such as title, price, condition, shipping cost, seller location, number of units sold, seller reviews, refurbished product status, and product URL.
… Features:
- Asynchronous scraping using Playwright
- Dynamic JavaScript rendering
- Proxy API integration
- Automatic pagination handling
- SQLite database for tracking page progress
- Automation of export to Google Sheets
- Continuation of scraping from saved page
- Extraction of structured product data
Technology stack:
- Python
- Playwright
- AsyncIO
- Requests
- SQLite
- Google Sheets API
- gspread
- dotenv
Workflow:
1. Load environment variables
2. Initialize SQLite database
3. Load rendered eBay pages via Scrape.do
4. Analyze product cards
5. Extract product data
6. Save products to Google Sheets
7. Save current page progress
8. Continue until all pages are fully scraped
The project is designed for scalable scraping and automation tasks in e-commerce.
-
18 USD Bypassing reCaptcha using a bot in Python
Bot Development
Bot for bypassing audio reCAPTCHA
An automation tool in Python that automatically solves Google reCAPTCHA v2 tasks — no manual data entry required.
… The bot launches a real Chromium browser, simulates human behavior with random delays, and bypasses bot detection mechanisms. When reCAPTCHA appears, it switches to the audio task, downloads the audio file, and transcribes it using Google Speech Recognition — then instantly submits the response.
Technology stack:
• Playwright — headless browser automation
• pydub + SpeechRecognition — audio processing and transcription
• Anti-detection: AutomationControlled flag disabled, spoofed User-Agent, navigator.webdriver overwritten
Features:
• Fully automated solving from start to finish
• Human-like interaction (random delays, real browser)
• Clean modular code — easily integrates into your project
• Tested on the official Google reCAPTCHA demo
-
17 USD Amazon Scraping
Data Parsing
Real-time integration with Google Sheets: uses gspread and Batch API to synchronize data directly from the cloud without exceeding request rate limits.
State preservation: built-in SQLite database to track progress, allowing the script to continue from the last page in case of interruption.
… Bot protection: implemented custom headers, User-Agents, and delays that mimic human behavior to ensure stable data extraction.
Automated data cleaning: cleaned and structured data, including URLs, titles, options, ratings, and prices.
-
17 USD Otodom scraping
Data Parsing
A reliable, ready-to-implement automation tool has been developed in Python, specifically designed for extracting large volumes of data from the Otodom.pl website. This project demonstrates a high level of competence in browser automation, database management, and data structuring.
Key technical features:
Constant data extraction state: an SQLite3 database backend has been integrated to track the progress of data extraction. This allows the script to remember the last processed page, ensuring instant recovery after interruptions — a critical feature for reliable data extraction.
…
Advanced browser automation: Playwright has been used to handle dynamic content, bypass cookie consent pop-ups, and simulate human interaction through smooth scrolling and random delays.
Clean data pipeline: automatically extracts and cleans complex fields, including title, price, price per m², area, number of rooms, and location.
Real-time export: a continuous data export system has been developed that saves results in Excel (.xlsx) after each page to prevent data loss.
Code quality: developed using an object-oriented programming (OOP) approach for maximum maintainability and scalability.
Technical stack:
Backend: Python
Automation: Playwright (Chromium)
-
16 USD Auto.ria scraping
Data Parsing
Successfully collected and structured data on the entire Ukrainian automotive market from the resource AUTO.RIA.
Key results:
Extracted detailed information on all car sale advertisements in the country.
… Structured complex raw data into a clear professional Excel table (XLSX) for market analysis.
Implemented high-speed asynchronous scraping to process large volumes of data.