Switch to English?
Yes
Переключитись на українську?
Так
Переключиться на русскую?
Да
Przełączyć się na polską?
Tak
  • Analysts of Pradesh

    Python
    The work is done on Python3+Qt5. Price Analyzer is the search for the item of the goods and the selection of the lowest price and delivery.


  • 68 USD

    Automation of marketing data collection (High-Speed Async Scrape)

    Python
    Development of an adaptive tool for data collection of any complexity. The system is designed to work with various types of web resources: from simple landing pages to multi-page catalogs and dynamic platforms. The main focus is on transforming unstructured HTML code into clean, business analytics-ready data.
    Technical capabilities (Senior level):
    Versatility of sources: Algorithms are configured to extract data from any structures: marketplaces, social networks, technical specifications, or news.
    High speed (AsyncIO): The use of non-blocking requests via aiohttp allows processing hundreds of pages simultaneously, minimizing wait time.
    Working with Big Data: Automatic serialization of results in JSON format, allowing easy import of data into databases, CRM systems, or Excel.
    Adaptive parsing: Use of BeautifulSoup4 for deep analysis of the DOM tree and extraction of specific meta-data (titles, prices, articles, descriptions).
    Mobile-Ready: The script is fully optimized for execution in the Termux environment, allowing parsing to run 24/7 directly from a mobile device or lightweight server.
    Result: The client receives not just a script, but a full-fledged data conveyor. All collected data is automatically saved in a structured file (parsed_data.json) in the device's internal memory for immediate use.


  • 113 USD

    AI bot infrastructure: RAG + Vector DB + Asyncio

    Python
    Demonstration of the core operation of my AI agent (real-time launch log).
    The screenshot shows the system initialization process:
    1. [SYSTEM]: The environment runs on Python 3.12.12 in asynchronous mode (Async/Await) — maximum request processing speed.
    2. [MEMORY]: Connection to the Pinecone vector database (index 'client-knowledge-base'). RAG mode (Active) is enabled, allowing the bot to respond accurately based on your documents rather than inventing facts.
    3. [NETWORK]: Checking connection to OpenAI API. A latency of only 142ms has been recorded — instant bot response.
    I am deploying this solution for clients who need a smart, fast, and stable bot for business.


  • 226 USD

    Mobile automation node (Termux Engine)

    Python
    In the screenshot is a demonstration of my working environment based on Termux. This is a mobile node that I use for traffic monitoring, quick debugging of scripts, and parsing in 24/7 mode.
    What is happening here:
    Stack: Working under Python 3.12.12.
    Automation: Running a custom bash script check_env.sh for instant checking of the environment's readiness for High-Load tasks.
    Emulation: A traffic monitoring and system status script has been implemented (Traffic: 842 KB/s).
    Cleanliness: Using professional command-line tools (awk, sed, cat) for on-the-fly data processing.


  • 428 USD

    Beryl AX traffic monitoring and management system (OpenW

    Python
    Development of an intelligent access control system for GL-iNet routers (Beryl AX). The project is implemented as a standalone solution, operating directly in the Termux environment on Android and interacting with the router in real-time via API.

    Technical challenges and solutions:
    - Compilation on ARM64: The pydantic-core kernel was successfully compiled on the Galaxy A36 mobile device, which required 100% CPU load and specific configuration of the system environment.
    - Smart Traffic Monitoring: An algorithm for filtering background traffic (threshold less than 50 KB/s) was implemented to accurately account only for the user's active screen time.
    - Asynchronous architecture: The use of the aiogram 3.x library allowed for instant bot interface response while continuously monitoring network activity in the background.

    Results (User Stories):
    - Dashboard: A professional interface with real-time usage limit status and quick session extension buttons.
    - Kill Switch: A function for instant complete blocking of internet access (Panic Mode) through direct commands to the router's firewall.
    - Automation: The system is deployed as a background service in a tmux session, ensuring stable operation 24/7 without the need to keep the terminal open.


  • 113 USD

    DataBridge Pro: Pricing automation and synchronization system

    Python
    Developed and implemented an autonomous system for e-commerce, automating the full cycle of managing the product catalog. The solution eliminates manual labor in recalculating prices and updating inventory from multiple suppliers.
    Tasks solved:
    Automatic data collection: Integration with external supplier APIs to obtain current price lists in real-time.
    Intelligent processing: Implemented flexible business logic: dynamic price recalculation considering currency fluctuations (USD/UAH) and customizable margins.
    Visualization and control: Created a professional console interface (CLI) with interactive tables and product status indicators (In Stock / Out of Stock) for operational monitoring.
    Reliability: The system is equipped with an operation logging module and automatic data backup to prevent losses during failures.
    Technology stack (Stack)
    Language: Python 3.12 (Core logic).
    Libraries: Rich (Professional CLI/UI), JSON, Requests.
    Database: SQLite (Local persistent storage).
    Automation/DevOps: Bash Scripting, Git/GitHub, Termux/Linux environment.
    Security: OAuth 2.0, Environment Variables (.env).


  • 113 USD

    Autonomous CRM system in Telegram with Google Sheets integration

    Python
    Task: Create a reliable lead collection system that operates 24/7, does not require monthly payment for CRM, and eliminates data loss during internet outages.
    What has been implemented:
    Hybrid database. Leads are instantly sent to Google Sheets (for manager convenience) and duplicated in a local database (for reliability).
    Data validation. The bot checks phone numbers, cleans them of unnecessary characters, and formats them uniformly.
    Admin panel. The owner manages the bot through a hidden menu: viewing statistics, checking server logs, and exporting the client database to a CSV file with one click.
    Fault tolerance. A Watchdog system has been implemented: if the bot process stops due to a server error, the script automatically restarts it.
    Protection against spam and queues. A finite state machine (FSM) has been implemented so that the user cannot break the bot's logic with chaotic clicks.
    Result: A fully autonomous tool for small businesses, replacing the manual labor of an administrator.
    Technology stack: Python, Google Sheets API, SQLite, Telegram Bot API.


  • CRM Telegram Bot

    Python
    **CRM system in the format of a Telegram bot + web admin panel for small businesses.** Management of contacts, deals, tasks, and documents directly from the messenger.
    ### Features
    * **Contacts and deals:** creation/editing/search, pagination. Sales funnel with stages, linking to contacts, history of changes.
    * **Tasks:** priorities, deadlines, linking to deals/contacts. Filter "mine/all", automatic reminders.
    * **AI/OCR:** photo of business card/invoice/receipt → OpenAI Vision extracts data → with 1 click a contact or deal is created.
    * **Reports:** export to Excel and PDF (support for Cyrillic), graphs. Integration with Google Sheets (export/import).
    * **Inline mode:** quick search for contacts and deals from any Telegram chat.
    * **Automation:** morning digest, task reminders, auto-cleaning of files.
    * **Web admin panel:** SQLAdmin with authorization for managing data through the browser + client documentation.
    * **Roles (4 levels):** User → Manager → Admin → Super Admin.
    * **Tracking:** Telegram notifications when visiting the admin panel and documentation.

    ### Stack
    Python 3.12, aiogram 3.24, FastAPI, SQLAlchemy 2 (async), Alembic, PostgreSQL 16, Redis 7, OpenAI Vision API, Docker Compose, Nginx, SSL, VPS.

    ### Security
    Data validation, rate limiting, connection pooling, protection against race conditions, security headers, non-root Docker, structured logging, centralized error handler.


  • 45 USD

    "Development of Telegram bots with AI integration (OpenAI/DeepSeek)."

    Python
    "Creation of multifunctional bots: from admin panels to complex systems using neural networks. Automation of data collection, website parsing, and working with databases (SQLite/PostgreSQL)."


  • 101 USD

    Script for quick purchase on the website

    Python
    Automated system for the purchase of numismatic products from the NBU online store.

    Professional Python script designed for the prompt acquisition of limited coins during peak load moments on the website.


  • 226 USD

    Markdown - Word/PDF Converter

    Python
    Production-ready solution for converting between Markdown and Word/PDF in both directions. Markdown → Word/PDF preserves LaTeX formulas (inline/display), code blocks, tables. Word/PDF → Markdown converts OMML formulas to LaTeX, tables to pipe format, extracts images. GitHub Flavored Markdown support out of the box.

    Desktop GUI on PyQt with drag & drop, two tabs for each direction, JSON formatting schemes for batch processing. Portable Pandoc included—zero dependencies setup. Libraries: mammoth for .docx parsing, pypandoc for conversion, custom regex for formulas.

    Handles complex cases: nested lists, tables in tables, mixed formulas. Settings persistence for recurring tasks.


  • 789 USD

    AeroTweet Farm

    Python
    A fully featured Twitter account farm automation system integrated with AdsPower and built for large-scale operations. It includes account management, group-based task distribution, custom action limits, automatic CAPTCHA solving, AI-powered text and image generation, profile behavior simulation, and adjustable content creativity. The system provides a spy module for competitor analysis, influencer tracking, shadowban detection, profile data monitoring, and automated ban detection. It also includes detailed statistics and logging, a Telegram bot for notifications and reports, and a user-friendly web panel for complete project control.