Messaging
Services that keep their promises when a queue backs up. Kafka, retries, dead letters, and ordering you can reason about.
Podgorica, Montenegro
Ivan Vukićević 👋
I build event-driven systems. Java, Spring Boot, Kafka, AWS.
Currently immersed in systematic, AI-assisted engineering.
What I’m good at. Four things, honestly assessed.
Cloud
Infrastructure that stays boring. AWS, containers, CI/CD — deploys that are a non-event.
AI-assisted engineering
Not autocomplete — a method. Claude Code, MCP, RAG, wired into planning, review, and testing.
Craft
Code the next person can change. TDD and real coverage, not a number on a dashboard.
Senior Software Engineer with almost 7 years of experience in enterprise software, currently immersed in AI-first development including Claude Code, MCP integrations, and RAG-based assistants with LLM integration. Core stack of Java, Spring Boot, Kafka, and AWS, with a focus on event-driven microservices and cloud-native workflows. Committed to clean, well-tested, maintainable code.
Latest writing. Things worth writing down.
All posts →Retrying Kafka messages without blocking the partitionA scheduler-backed retry topic keeps transient failures from stalling everything behind them.Exactly-once is a contract, not a checkboxKafka's transactional producer gets you atomic writes. Everything either side of the broker is still your problem.What actually fixed our Lambda cold startsProvisioned concurrency is the expensive answer. Three cheaper changes moved p99 further.