Nowadays everyone is talking about AI. On Threads, a lot of programmers brag about their vibe-coding agents. They build end-to-end systems only by using prompts and markdown files to tell AI the rules they should follow. I’m not an anti-vibe-coding person because I’ve tried it too, building things with agents and MCP. But in the end, I feel it’s not my cup of tea.

What I prefer is to use AI when I really need it, like when I’m trying to answer questions that I’d otherwise search for on Stack Overflow, cases where there are no clues from my code linter, application logs, or system logs.

I used to code using Neovim, and there’s a plugin called CodeCompanion. You have to check out their website, it’s an awesome plugin! It has agentic, MCP, and other nice features, but honestly, I never use those features anymore except for the editor context: #buffer #buffers #diagnostics.

What I do with CodeCompanion is as easy as asking about what, why, and how, not generating whole chunks of code that could burn through all your tokens. I usually only need $6 of Claude API credit per month. I don’t know about yours, but I feel this fits my needs.