AUTOMATE · 35 ARTICLES
Continue through the latest AI coding and agents articles.
All articles
When red error messages cover the screen, your heart sinks. The English is dense, full of unfamiliar words, and seems not to tell you what went wrong. So many vibe coders close the message without reading it and tell AI, "It doesn't work—fix it." Then AI starts changing the wrong parts.
Read articleThe app is complete. It runs perfectly on my computer. I copy localhost:3000 from the address bar to show it off to a friend, but they reply, "It won't open." I finally deploy it, only to find that the app that worked locally now throws errors on the internet.
Read articleThe benefit of delegating work to AI coding agents is that you do not need to stay at your desk. But leaving your desk immediately creates another problem. If an agent stops after asking a question, the entire task waits until you return. You end up unable to leave your laptop—the paradox of delegation…
Read articleVibe coding inevitably brings this day. You ask AI to add one feature to an app that worked fine yesterday, and it changes things everywhere until the entire app stops working. Even if you ask, "Put it back the way it was," AI cannot restore it precisely. It changed twenty places, and it does not know which one is broken…
Read articleIn Part 1, I compared an app to a restaurant: dining area (frontend), kitchen (backend), and storage (DB). This time, we cover the topic most likely to cause trouble in vibe coding: APIs—and API keys. Everyone has seen “Never expose your key,” but few explain what the key is, where it lives, or what happens when it leaks. This article covers it all.
Read articleWhen using AI coding tools, you often repeat the same instructions: “Use this format for commit messages” or “Follow this checklist before deployment.”
Read articleWhen using multiple AI coding agents, you may spend more time checking “who is doing what on which branch” than writing code. With several terminals open, branches changing, and diffs to compare, the benefits of parallel work quickly fade.
Read articleMore people are building apps by describing them to AI. Tell tools like Cursor, Claude Code, or v0, "Build this service," and a working app appears. But after building it, many hit the same wall: the app runs, yet they do not know what AI actually created…
Read articleIn earlier installments, we explored ways to conserve context: clear history at each work boundary (Part 3) and keep fixed overhead short (Part 4). Yet some tasks remain too large to handle. A typical example is, “Figure out how payment logic flows through this codebase.” A diligent agent would read dozens of files, and as we saw in Part 1, everything it reads accumulates in context. By the time the investigation is complete, no context remains for the actual work that knowledge enables—changing the code.
Read articleThe final part of the series is about money. If you have run an agent through an API, you may have noticed something odd on the bill. Input tokens cost vastly more than output tokens. Given the architecture from Part 1, that is only natural. Every turn resends everything from the system prompt through the full conversation history, so in a 50-turn agent session, the same system prompt is effectively billed 50 times…
Read articleIf you have followed the series this far, you will see a common pattern: minimize what stays inside the context window. Part 3’s “write state to a file before /clear,” Part 4’s “only pointers instead of the body,” and Part 5’s “leave the process to subagents, send only conclusions to the main agent” all point in the same direction. But we have not yet properly addressed the destination of these files—that is, information sent outside the context.
Read articleIn Part 3, we saw how to clear conversation history with /clear. But checking the remaining context immediately afterward shows it is not at 100%. Run /context in Claude Code and the reason becomes clear: the system prompt, tool definitions, CLAUDE.md, and tools registered by MCP servers already consume tens of thousands of tokens…
Read article