— PORTFOLIO

SENIOR AI ENGINEER

I build multi-agent AI systems that ship to production.

01 — ABOUT

Background & Approach

I spent 6 years building systems that don't break — enterprise Java, Spring Boot, distributed microservices at GlobalLogic. That foundation is why my AI systems are different: they have durable event queues, per-tenant data isolation, real observability, and zero event loss across restarts. Not because those are nice to have, but because I come from a world where those are non-negotiable.

Today I ship agentic systems to production. At GlobalLogic I built an agentic CI system (Claude Code + LiteLLM) that autonomously integrates upstream open-source commits with human-in-the-loop guardrails — and resolved a platform upgrade stalled for 22 months. On my own platforms I implement the protocols, not just the frameworks: custom MCP HTTP servers with Streamable HTTP transport and RFC 9728 OAuth discovery, full A2A systems with task lifecycle management, SSE streaming and multi-turn flows, and LangGraph orchestration over per-user ChromaDB RAG with strict tenant isolation.

CURRENTLY SEEKING

I'm looking for a product company where the AI architecture is the product — not a thin wrapper around an API call. I thrive designing multi-agent orchestration systems, building MCP servers, and working with teams that treat observability and fault tolerance as first-class concerns, not afterthoughts. Remote only. Contract or full-time.

02 — SKILLS

Technical Stack

Protocol Engineering

MCP (Streamable HTTP, RFC 9728 OAuth discovery), A2A (task lifecycle, SSE streaming, multi-turn input-required flows), BYOK architecture, multi-tenant RAG isolation

Agentic Systems

LangGraph stateful orchestration, Claude Code, LiteLLM, durable Redis event queues, per-user ChromaDB isolation, shared fallback collections, zero event loss design, n8n workflow automation

LLM Integration

Anthropic Claude, OpenAI API, Google Gemini, Groq, Ollama, OpenRouter · multi-model fallback · token cost optimization · structured output · prompt engineering · evaluation harnesses

RAG & Knowledge

ChromaDB, Pinecone, pgvector, FAISS, Neo4j · GraphRAG · embedding optimization · chunking strategies · contextual retrieval · semantic caching (Redis)

Backend Foundation

Python (FastAPI, Pydantic) · Java 17 (Spring Boot 3, Spring Cloud) · Go · PostgreSQL · Redis · Kafka · event-driven architecture · microservices

Infrastructure

Docker · Kubernetes · AWS (EC2, RDS, S3, EKS) · Terraform · Ansible · GitHub Actions CI/CD · nginx · Cloudflare · OVHCloud VPS

Observability

Prometheus · Grafana · OpenTelemetry · cAdvisor · audit trails · GDPR compliance (right to erasure, data export, retention automation)

Certifications

Neo4j Certified Professional · 10× Anthropic (Claude Code in Action, Building with the Claude API, Claude in Amazon Bedrock, Claude with Vertex AI, Subagents, Agent Skills, and more)

03 — PROJECTS

Selected Work

01Dec 2025 — Present

Multi-Agent SDLC Orchestration Platform

GitHub automation powered by three specialized LangGraph agents on per-user ChromaDB RAG knowledge bases. Stateful workflow graphs with durable Redis event queues — zero event loss across restarts — and full Prometheus/Grafana observability.

LangGraphFastAPIChromaDBRedisPostgreSQLPrometheus/GrafanaGitHub Actions
02Feb 2026 — Present

Multi-Protocol AI Agent System (MCP + A2A)

Clinical prior-authorization AI running three interoperability protocols together — SMART on FHIR for data access, MCP for LLM tool interfaces, and A2A for inter-agent delegation. Custom MCP HTTP server with 8 tools, connectable directly from Claude.ai; full task lifecycle with SSE streaming and multi-turn flows.

FastAPIMCP Streamable HTTPA2A ProtocolFHIR R4SSEOpenRouterDocker
03Sep 2024 — Present

AI-Powered Trip Cost Platform

Production trip-cost calculator with a LangGraph route-planning agent — natural-language route creation and modification against the live map — plus shareable trip receipts with server-injected Open Graph previews. Multi-provider routing (Mapbox with OSRM fallback), seasonal glassmorphism design system, Spring Boot single-JAR deploy.

ReactTypeScriptSpring Boot 3LangGraphMapbox GLRedisPostgreSQLDocker
04Apr 2024 — Present

Production n8n Automation Platform

Self-hosted n8n in queue mode on OVHCloud, backed by PostgreSQL and Redis. Powers Telegram bots, agentic AI workflows across OpenAI/Claude/Groq, custom MCP servers, and cross-service automation across all active projects.

n8nPostgreSQLRedisOVHCloudDockerMCP
05Jul 2025

Banking REST API in Go

Account management and money transfer API with ACID-compliant transactions and deadlock-safe concurrent transfers. Type-safe database layer generated with SQLC, PASETO token authentication, 96% test coverage, deployed on AWS EKS through CI/CD.

GoPostgreSQLSQLCPASETODockerAWS EKSGitHub Actions
06Feb 2025 — Present

RAG Automotive Diagnostics Assistant

RAG pipeline indexing technical PDF manuals in Pinecone with Ollama embeddings. A Telegram bot fronts it with LangChain for natural-language diagnostic queries, with self-hosted LLM support for offline operation.

RAGLangChainPineconeOllamaTelegram Bot APIPython

04 — CONTACT

Let's Talk

Interested in AI systems, agentic platforms, or consulting? Drop a message — I read everything.

Based in Ukraine. Available for remote contracts, consulting, and full-time roles. Response time: 24–48h.