Before working for 2 years on the Apache APISIX API gateway, I was mainly oblivious to API gateways. It’s only by working with them that I understood their value. Decoupling the client and the server unlocks a lot of options: moving authentication to the API Gateway, securing APIs, deduplicating API requests, etc. In this post, I want to describe how the same pattern applies to AI. AI gateways AI gateways work in a similar way.

I like to analyze codebases I start working on, or that I left for months. I ask my coding assistant, case in point, Copilot CLI: 'analyze the following codebase and report to me improvements and possible bugs.' It’s vague enough to leave room for crappy feedback, but also for some interesting insights. I did it last week on a code base. Copilot returned a list of a dozen items. I asked it to create a GitHub issue for each, with the relevant labels, including priority.

Most developers now use coding assistants. I do too—Copilot at work, Claude Code at home. As a developer, I prefer not to repeat myself. This post explains why and how to avoid repetition as a skill. Don’t Repeat Yourself The DRY principle has been present in the software development field for ages. The idea is that if you copy and paste code in multiple places and a bug appears, you’ll need to fix the bug in all these places.

Last year, I started to dig a bit around LangChain4J. It’s a fast growing project, and I wanted to get familiar with the updates. I also wanted to check how I could integrate a Model Context Protocol server in LangChain4J.

I’m coming relatively late to the LLM party, but I rarely come very early in the hype cycle. For example, I never bought into blockchain, the solution still searching for problems to solve, nor in microservices, the latest in the cargo cult IT trends. Despite my late arrival at the LLM party, I have been a regular user of LLMs. I use OpenAI for non-controversial questions outside my cone of knowledge, e.g., linguistics or legal; I use GitHub Copilot in my IDE to improve my code.