• Location: Kyiv, Ukraine
• Projects Portfolio: https://apps.bogdanovych.org/en/

Interactive Flet-based chat interface featuring configurable historical and fictional Ukrainian characters powered by Lapathoniia LLM provider.

• Architected and developed a cross-platform (Web & Desktop) conversational application using Python 3.12 and Flet (Flutter-based framework).
• Engineered dynamic character customization allowing rapid onboarding of new personas (such as Taras Shevchenko, Lesya Ukrainka, and Carpathian Hutsul) via YAML-based system prompt configurations.
• Integrated asynchronous LLM communication utilizing the openai client (AsyncOpenAI) optimized for the Lapathoniia API, featuring user-controlled model parameters (temperature, max tokens).
• Implemented reliable state and configuration management powered by Pydantic v2 and Pydantic Settings.
• Streamlined deployment & development workflow using Docker and uv for fast package synchronization and reproducible execution environments.

GitHub: https://github.com/BogdanovychA/slovohr-ai
Web: https://slovohr-ai.bogdanovych.org/

Open-source hotline automation system for government sector digitalization.

• Architected and developed a cross-platform application using Python 3.14 and Flet, strictly adhering to SOLID principles and modular design for long-term maintainability.
• Automated citizen appeal processes for the State Energy Supervision of Ukraine, implementing a scalable document generation engine (DOCX, MD, HTML) using the Registry Pattern and Jinja2 templates.
• Engineered a robust communications layer with secure SMTP (SSL/TLS) delivery and enterprise-grade data validation/configuration management using Pydantic v2.
• Streamlined deployment and development workflows using Docker and uv, ensuring consistent execution across Desktop, Web, and server-side environments.
• Standardized for Open Source & Research: Orchestrated the public release under EUPL-1.2, ensuring full metadata compliance with PEP 639 and establishing research citability via DOI.

GitHub: https://github.com/BogdanovychA/h-line

Asynchronous data persistence library for Flet applications.

• Designed and developed a reusable storage solution with automatic JSON serialization for complex data types
• Implemented data isolation via namespacing, enabling safe multi-module usage
• Published and actively maintained as an open-source package on PyPI

GitHub: https://github.com/BogdanovychA/flet-storage
PyPI: https://pypi.org/project/flet-storage/
AI Skill: https://skills.sh/bogdanovycha/flet-storage/flet-storage

Lightweight Project Fluent localisation manager with automatic locale fallback.

• Designed and developed a streamlined localization solution featuring automatic discovery of .ftl resource files from the filesystem
• Implemented a robust fallback chain (preferred locales → default locale → key itself) to ensure reliable string rendering across multiple languages
• Published and actively maintained as an open-source package on PyPI

GitHub: https://github.com/BogdanovychA/fluent-manager
PyPI: https://pypi.org/project/fluent-manager/
AI Skill: https://skills.sh/bogdanovycha/fluent-manager/fluent-manager

Lightweight Python wrapper for reliable retrieval of YouTube transcripts with automatic URL parsing and subtitle fallback.

• Designed and developed an open-source Python library that simplifies transcript extraction from YouTube videos across multiple URL formats, including youtube.com, youtu.be, shorts, and embed
• Implemented a resilient transcript lookup flow that prioritizes manually created subtitles and automatically falls back to YouTube-generated transcripts when needed
• Added developer-friendly features such as language discovery, optional timecoded output, and type hints to improve integration into automation, research, and content-processing workflows

GitHub: https://github.com/BogdanovychA/mr-transcript
PyPI: https://pypi.org/project/mr-transcript/
AI Skill: https://skills.sh/bogdanovycha/mr-transcript/mr-transcript

Lightweight asynchronous Python client for the Google Analytics 4 (GA4) Measurement Protocol.

• Designed and developed an asynchronous client using httpx to send telemetry and events to Google Analytics 4.
• Implemented connection pooling with async context managers and support for custom httpx.AsyncClient injection to optimize network overhead and reuse TCP connections.
• Integrated a debug/validation mode using GA4's validation server (/debug/mp/collect) to verify event payload structures without logging dummy production events.
• Structured support for complex parameters, including nested e-commerce tracking objects (items), lists, dictionaries, and primitives.

GitHub: https://github.com/BogdanovychA/measurement-api
PyPI: https://pypi.org/project/measurement-api/
AI Skill: https://skills.sh/bogdanovycha/measurement-api/measurement-api

Telegram-based interface for interacting with multiple LLM providers.

• Designed secure API key storage system using AES encryption
• Enabled safe multi-user interaction with Gemini, ChatGPT, and Claude
• Implemented multilingual support

GitHub: https://github.com/BogdanovychA/nexus-shell
Bot: https://t.me/NexusShellBot

Social application to honor the memory of fallen heroes.

• Developed cross-platform application (Web, Android) using Flet
• Implemented multilingual support
• Delivered a user-facing product with real-world purpose

GitHub: https://github.com/BogdanovychA/CodeUA
Web: https://codeua.foundation101.org/
Google Play: https://play.google.com/store/apps/details?id=org.foundation101.codeua

Collection of casual games for mobile and web platforms.

• Developed cross-platform applications using Flet
• Built and deployed web and Android versions
• Structured multi-project architecture for scalability

GitHub: https://github.com/BogdanovychA/bmg
Web: https://minigames.bogdanovych.org/
Google Play: https://play.google.com/store/apps/details?id=org.foundation101.minigames

RPG game based on D&D 5e mechanics with custom Ukrainian content.

• Developed backend logic for combat system, progression, and dice mechanics
• Integrated LLM APIs for dynamic interactions
• Built web-based interface using Streamlit and FastAPI

GitHub: https://github.com/BogdanovychA/karatel-game
Web: https://karatel.ua/