Kiryl Rusanau

I have production experience integrating LLM capabilities into enterprise Java applications using LangChain4j and Spring AI frameworks. This includes building RAG (Retrieval-Augmented Generation) pipelines with vector databases, implementing prompt chains for structured output parsing, and deploying conversational AI systems. My AI work spans both enterprise integrations and consumer-facing AI products like CartSync.

My primary AI toolkit includes LangChain4j for Java-based LLM integration, Spring AI for enterprise Spring Boot applications, and LangGraph (Python) for multi-agent workflow orchestration. I work with various LLM providers and have experience implementing vector stores, embeddings pipelines, and production-grade prompt engineering patterns.

I follow a structured approach: designing RAG architectures for domain-specific knowledge retrieval, implementing vector embeddings with appropriate chunking strategies, building prompt chains that produce reliable structured outputs, and ensuring proper error handling and fallback mechanisms. The integration respects enterprise concerns like security, observability, and cost optimization.

I've built CartSync, a conversational AI grocery planning assistant on the Telegram platform that uses natural language processing for intelligent list management and shopping suggestions. I'm also developing Quibench, an AI-powered productivity platform that leverages LLM capabilities for workflow optimization and team collaboration.

CartSync is an AI-powered grocery planning assistant available as a Telegram bot. It uses conversational AI to help users manage shopping lists intelligently, provides smart product suggestions based on context, and learns from user preferences. Built with a Java backend, it demonstrates practical AI integration in consumer applications.

Quibench is an AI productivity platform I'm developing that focuses on workflow optimization using LLM technologies. It aims to enhance team collaboration and individual productivity through intelligent automation and AI-assisted task management.

I have 7+ years of experience as a full-stack engineer with deep expertise in Java (8-21), Spring Boot ecosystem, React with TypeScript, and AWS cloud infrastructure. My background includes building microservices at scale, implementing OAuth2/OIDC authentication systems, and working with event-driven architectures using Kafka.

I spent nearly 4 years at Azati Corporation working on financial technology projects including cryptocurrency custody platforms with Fireblocks integration, securities registration systems for regulatory compliance, and banking authentication systems managing identity across multiple services. This experience shaped my understanding of security-first development and compliance requirements.

Yes, I'm available for consulting engagements focused on AI/LLM integration, particularly LangChain4j and Spring AI implementations, RAG architecture design, and enterprise Java development. I also take on technical advisory roles for startups building AI-powered products.

I'm based in Poland and work with clients worldwide. The best way to reach me is via email at [email protected] or through LinkedIn. I'm open to both remote collaboration and on-site engagements within Europe.