Юрій

Yurii Bubelnyk
AI Engineer · LLM Systems
SUMMARY
AI Software Engineer with hands-on experience designing LLM-based applications, RAG
architectures, and multi-agent systems. Skilled in semantic retrieval, fine-tuning, and
integrating AI into production-grade backend services using LangChain, LangGraph, and
vector databases. Focused on scalable AI systems, MLOps practices, and business-focused
automation.
EXPERIENCE
BeliefCraft — SoftServe Academy 5 months
Yurii Bubelnyk Backend & Agent Systems Engineer
AI Engineer · LLM Systems › Built and deployed a production multi-agent RAG system for decision-making in partially
observable environments, optimizing latency and reliability.
CONTACT › Implemented backend services with LangChain and LangGraph for agent orchestration;
developed an Environment Sub-Agent for structured data access via FastAPI.
[відкрити контакти](див. вище в блоці «контактна інформація»)
› Debugged agent-service integrations and maintained technical documentation.
[відкрити контакти](див. вище в блоці «контактна інформація») Python · FastAPI · LangChain · LangGraph · Weaviate · PostgreSQL · LangSmith · Docker
Lviv, Ukraine
github.com/Yachiry1
AI Movie Recommendation Platform 9 months
Python Backend Developer
linkedin: yura-bubelnyk
› Developed backend for a recommendation system using FastAPI with semantic search via
AI / GENAI embeddings and pgvector.
› Designed and optimized PostgreSQL schema; improved recommendation relevance and query
RAG · LLM Systems performance.
LangChain · LangGraph Python · FastAPI · PostgreSQL · Qdrant · SentenceTransformers · Docker
LangSmith · LlamaIndex
Multimodal Cocktail RAG System Pet project
Multi-Agent Systems
AI / RAG Developer
CLIP · Reranking
› Built a multimodal RAG recommendation system using text, images, moods, and colors with
BACKEND & DBS CLIP embeddings for image-text similarity matching.
› Improved recommendation quality through retrieval evaluation and reranking experiments.
Python · FastAPI · Django Python · RAG · CLIP · Vector Search · Reranking
REST APIs · Celery
PostgreSQL · MongoDB
Skyvern MCP Directory Research 9 months
AI Tooling / MCP Research Contributor
Qdrant · FAISS
› Researched MCP directory platforms and AI agent tooling ecosystems; prepared structured
pgvector · Weaviate listing content and technical metadata for MCP server discovery platforms.
ChromaDB MCP · AI Agents · Technical Research · Metadata Structuring

DEVOPS & OTHER CORE STRENGTHS
Docker · CI/CD › Strong focus on LLM applications, RAG pipelines, multi-agent workflows, backend integration,
MLOps · Git · C++ and debugging production-oriented AI systems.
› Comfortable working across APIs, databases, vector search, Docker-based environments,
EDUCATION documentation, and technical research.
Bachelor's in Computer Science
Lviv Polytechnic National University
Expected 2028

CERTIFICATE
Multi-Agent RAG Systems
& Intelligence Engineering
2026