Specialization: Anti-bot bypass systems, high-load Telegram ecosystems, production-grade web automation
Impact: 9,000+ active users across bot infrastructure | 2+ years production experience
🟢 Available for remote opportunities | Mid-level Backend/Automation Engineer

Python Backend Engineer with 2 years of commercial experience in automation systems and data extraction.

I solve complex engineering challenges:

• Anti-bot systems: Bypassing Cloudflare/Incapsula with 99% success rate
• High-load architecture: Building fault-tolerant bot ecosystems serving 9,000+ users
• Production scraping: Self-healing ETL pipelines processing 1,000+ records/day

Background: Ex-Team Lead with leadership experience, ensuring maintainable and business-oriented code.

Core: Python 3.10+, AsyncIO, OOP
Automation: SeleniumBase (UC Mode), Playwright, Selenium, Appium
Bots & APIs: Aiogram 3, Telethon, FastAPI, python-telegram-bot
Infrastructure: Docker, Linux (Systemd), MongoDB, PostgreSQL
Data Processing: Pandas, Google Sheets API, JSON/XML, PyPDF2
AI Integration: OpenAI API, LangChain, LangGraph, Gemini API
Tools: Git, Postman, Chrome DevTools (Network Analysis)

Problem: Automated registration blocked by Cloudflare and Incapsula
Solution: Custom Chrome Extension for proxy authentication + human behavior emulation
Impact: 99% success rate on React SPAs with dynamic protection
Tech: SeleniumBase (UC Mode), Manifest V3, RuCaptcha API, Proxy rotation

Problem: Manual job posting distribution across 20+ Telegram channels
Solution: Cluster of 3 interconnected bots with fuzzy-matching algorithm
Impact: 9,000+ active users | 80%+ filtering accuracy | Zero downtime
Tech: Python, Aiogram 3, Telethon, MongoDB, Docker, systemd

Problem: Extracting vehicle auction data from protected websites and PDFs
Solution: Multi-stage ETL with AI extraction, VIN validation, and mobile browser emulation
Impact: Automated processing of 1,000+ records/day | Bypassed CAPTCHA protection
Tech: Python, Gemini API, Playwright, Appium, NHTSA VPIC API, Google Sheets

Problem: Period tracking apps sell user data to advertisers
Solution: End-to-end private Telegram bot with partner linking and medical reporting
Impact: Production deployment | Full cycle product development
Tech: Python, python-telegram-bot, MongoDB, Docker, pytz, JobQueue

Problem: Users need unified tool for downloading from Instagram, TikTok, X, LinkedIn
Solution: Modular handler architecture with smart fallback chain (yt-dlp → API → metadata)
Impact: Active production bot with media group support and broadcast system
Tech: Python, pyTelegramBotAPI, Instaloader, yt-dlp, Selenium

Problem: Need stateless AI assistant with function execution
Solution: LangGraph state machine with OpenAI tool integration
Impact: Foundation for AI-assisted automation tools
Tech: Python, LangGraph, LangChain, OpenAI GPT-4o, Pydantic

✅ Production mindset: Every project designed for fault tolerance, logging, and restart safety
✅ Anti-bot expertise: Deep knowledge of browser automation, CAPTCHA solving, and proxy management
✅ Full-stack approach: From Selenium scripts to Dockerized microservices with MongoDB
✅ Technical leadership: 5+ years managing teams and documenting complex systems
✅ Business-oriented: Focus on metrics, uptime, and scalable architecture

💼 LinkedIn: linkedin.com/in/helena-gusarevich
📧 Email: Pdf.man.3d@gmail.com
💬 Telegram: @Hel_inst
🌐 Portfolio: hel-3d.github.io/hel-3d
🌍 Location: Yerevan, Armenia (UTC+4) | Open to relocation

Building production-grade automation systems that never sleep