Роман

Roman Nalapko | Python Developer • Web Automation • Scraping • ETL Pipelines
Email: [открыть контакты](см. выше в блоке «контактная информация») | Telegram: @BEATZIRO | GitHub: https://github.com/yourprofile |
LinkedIn: [открыть контакты](см. выше в блоке «контактная информация») | Location: Romania / Ukraine
(Remote) | English: B1

Summary:
Python Developer focused on web automation, scraping, ETL pipelines and high-load data
processing. Experienced in building scalable async/multithreaded applications, bypassing anti-bot
systems, working with proxy/session management, and deploying production-ready desktop &
backend tools. Skilled in Docker, Linux, Playwright, Telegram/Instagram automation, API integration
and performance optimization.

Technical Skills:
Python: asyncio, aiohttp, requests, BeautifulSoup, lxml, pandas ​
Automation & Scraping: Playwright-stealth, proxies, session management, bot-detection bypass ​
Desktop: PyQt5/6, threading, workers, MVC ​
Databases: PostgreSQL, MySQL, SQLAlchemy
DevOps: Docker, Docker Compose, Linux ​
Concurrency: multiprocessing, multithreading, async pipelines ​
Other: API integration, captcha handling, PyInstaller

Projects:
ChipDale Telegram Bot — 2022
Telegram bot MVP for a local on-demand services startup, automating the entire task request flow
(cleaning, moving, repairs, delivery).
Technologies: Python, pyTelegramBotAPI, OOP, Templates, inline/reply keyboards, multi-step form
logic. Implemented structured request processing, user data collection, message templating, and
automatic forwarding to the operator channel.​
Result: Reduced request creation time from 5–7 minutes to 40 seconds; increased daily processed
tasks up to 100; decreased operator workload by 60%.
Commercial Website — skliarova.com — 2023
Client website built together with a designer, focused on UX structure, adaptive layout, and brand
presentation.
Technologies: Wix, SEO Tools, forms, integrations, responsive layout.
Implemented full page structure, content hierarchy, mobile/tablet/desktop optimization, basic SEO
and tracking.
Result: Increased conversion by 12–18% and boosted organic traffic by 15–30% after launch.
LuxWash Assistant Bot — 2024
Telegram/SmartSender assistant for a self-service car wash managing real-time queue, feedback,
and customer communication.
Technologies: SmartSender flows/triggers, message blocks, Telegram integration. Implemented
automated menu navigation, queue display, feedback submission, location sharing and promotional
info delivery.
Result: Reduced incoming calls by 50–65%, increased repeat customers by ~20%, and tripled
customer feedback volume.
Instagram Auto-Poster (Telegram + Make.com + Dropbox) — 2025
Automation system for preparing Instagram content via Telegram bot with automated media storage
in Dropbox through Make.com scenarios.
Technologies: Make.com (webhooks, JSON processing), Telegram Bot API, Dropbox integration.
Implemented Telegram → JSON → Make pipeline, automatic content structuring, file uploading, and
templated metadata handling.
Result: Reduced content preparation time from 5–10 minutes to 10–15 seconds, saving up to 20
hours/month for SMM teams.
Speech-to-Text Telegram Bot (Open Source) — 2025
Telegram bot converting voice/audio/video messages into text using faster-whisper with optional .srt
subtitle generation.
Technologies: aiogram 3.7+, faster-whisper, FFmpeg, asyncio, ThreadPoolExecutor, SRT generator,
Python 3.9+. Implemented 16kHz WAV conversion, Whisper transcription, auto language detection,
SRT output, model configuration commands and parallel processing.
Result: Transcribes 3–5 minute audio in 1–2 seconds; speeds up note-taking by 5–8×; reduces
manual typing by up to 90%.
Instagram Data Collector & Scraper — 2025
Desktop application for authorized Instagram content collection (Reels/Post) with date filtering, used
by marketers, content creators, and analysts to automate data collection.
Technologies: Playwright / Playwright-stealth, Requests, aiohttp, BeautifulSoup, lxml,
cookies/session handling, bot detection bypass (headers, proxies, delays), multiprocessing, ETL
pipelines, XLSX export. Implemented scaling via ProcessPoolExecutor, error handling, logging,
deduplication, and retry mechanisms.
Result: Processes 300+ profiles/hour, reducing manual workload for analysts by over 90%.
Telegram Automation Toolkit — 2025
Desktop application for parsing Telegram groups and automated invitations, supporting audience
analysis and growth strategies.
Technologies: PyQt6, Telethon, asyncio, QThread, .session files, SOCKS5 proxy support, anti-spam
logic, import/export of participants. Asynchronous processing with error handling and logging,
continuation after interruptions, smooth automated workflow.
Result: Automates audience collection up to 1000+ users/hour, saving 40+ hours/month for SMM
teams.

Education
Navigation Student (3rd year) — Currently studying