Edit Article

I get teased a lot for using AI in most of my communication. Some people say relying on AI will eventually make me lazy. Some assume that if I let an LLM draft a technical note, it means I am losing the ability to reason, write, or build systems on my own. My answer is simple: I have never been ashamed of using AI, and that includes my technical articles and system documentation.

For me, AI does not replace my thinking. It helps me organize the thoughts that are already running through my head. Ever since I started working with AI, IoT, and neural systems, I have had too many things I wanted to share: too many edge cases, too many sensor behaviors, too many API contract quirks, too many failure modes, and too many moving parts I feel are important to explain.
When I start writing, I can go for hours. A small integration bug can easily turn into a long explanation because I see the code path, the production context behind it, the network topology, and the wider architecture. In team chats, some people have joked about the 'Agent TC is typing...' status because I take so long to compose a message. Honestly, I take that as a compliment.

The problem was never a lack of understanding. The problem was having too many layers to structure and not enough time to lay them out cleanly. Sometimes, by the time I finally hit send, the incident had already been mitigated or the thread had gone cold. That is exactly why AI became useful for someone like me.

AI does not write on my behalf. It helps me restructure what I already know. Most of the time, my original prompt is longer than the final version the model produces. The reasoning, direction, and substance still come from me. AI just makes them clearer, sharper, and easier for the rest of the team to act on.

To me, using AI is not a sign of laziness. What is truly wasteful is having hard-won knowledge, production experience, design lessons, and debugging stories, but never sharing them because the writing process feels too heavy. AI is just a tool, like a compiler for code, a debugger for state, or an observability stack for production signals. A good tool does not make an engineer weak. It helps an engineer ship better systems.

Today, at AINNA, I use AI not only for writing, but also for building the detached systems that support my daily work. From technical docs and incident updates to operational pipelines, telemetry processing, and decision-making workflows, AI has become part of how I design, deploy, and maintain systems. So people can say whatever they want. I will keep using AI, not because I want to avoid thinking, but because I want my thinking to reach the systems and the people who build them more clearly.