About

I have worked on 24 projects over 18 years and 6 months, including roughly 15 years of hands-on experience with Java. From two-person development to a 500-person system overhaul, across more than 10 industries — finance, telecom, logistics, insurance, healthcare, broadcasting, security, leasing, e-commerce, travel, and public sector — I have been involved in every phase of development: research and analysis, requirements definition, basic design, detailed design, implementation, testing, infrastructure setup, and operations. I started as a programmer, then grew into SE and lead roles, and in recent years I have increasingly been brought in from the requirements definition stage.

What I am doing now

Lately, in my work, I have been building agents (sets of prompts that direct an AI) that take a design document, add and modify source code, produce test specifications, generate JUnit test classes, and run unit tests — and I keep tuning them until they are accurate enough to use on a real project. It is not only the wording of the instructions: how tightly you narrow down the information you feed the AI (the context) also drives accuracy and speed, so I tune that as well. Through this, I have come to see where the line falls between what can be delegated and what cannot.

Simple coding and testing are areas where I expect AI to take on more and more. What I think will remain with people is judgment grounded in an understanding of the business itself. The reason I am building AI agents into real work right now is to find out for myself exactly how much can be handed over.

What I can do

From front end to back end to infrastructure, I have handled the full range within the same project.

As tech lead on a large-scale system overhaul, I drove the migration from Struts to Spring Boot 3 and from JSP to Vue.js / TypeScript.
I have worked on analyzing legacy assets and replatforming them onto modern environments (UNIX to Linux migration, consolidating multiple servers into the cloud, and so on).
I have handled a great many difficult bug investigations and incidents in production environments. I isolate the cause from both log analysis and source code analysis.
I join from the requirements definition stage, and I value the process of shaping functional specifications and migration plans through dialogue.
As a lead, I work from WBS-based task management and thorough code review, through to building DB tuning and security measures (SQL injection, XSS, and so on) into the design stage itself.

What is written here is only a small part of the technical side. My résumé covers which problems I faced, and how I approached them, across more than 10 industries.

What I am building

Alongside my professional work, I keep building systems on my own.

A personal RAG system (retrieve, then assemble an answer) combining ChromaDB with a local LLM
A platform combining DVD rental and video streaming (admin console, customer-facing app, mobile app)
Full-stack development of a microservices + BFF (a relay API dedicated to the front end) architecture handling money movement such as bank deposits, withdrawals, and transfers
An electric scooter sharing service

From planning through design, implementation, and operations, I see it through with my own hands rather than leaving it on paper — I make it actually run.

How I approach writing

What I care about when writing a technical article is conveying the whole picture of how a system takes shape and comes to life, rather than presenting disconnected techniques. I back my claims with reasoning, and I want to build relationships where the people reading and I sharpen each other technically. The passion I felt when I first started in this work is still with me, unchanged.

Contact