AI Code of Conduct

As a developer, I use AI daily — including building LLM and RAG integrations for clients. I also believe engineering judgment isn't something to hand off. Here's where I draw the line.

AI is a genuine part of how I work: it helps me move faster through research, catch issues earlier, and handle tedious groundwork that used to eat hours. It's also something I build with — integrating LLM APIs and retrieval systems is part of my skill set, not just a tool I use on the side.

But I don't let it replace understanding. Every line of code I ship, I've read, reviewed, and can explain — whether I wrote it or a tool helped draft it. AI accelerates my work; it doesn't do the work in my place.

So here's my promise: I stay hands-on-keyboard on the decisions that matter — architecture, security, and anything where a wrong assumption is expensive. I use AI to augment my judgment, not to substitute for it. And if a client ever wants to know where and how AI was used on their project, I'll tell them plainly.

Questions?

Happy to talk through any of this — reach out at nikola@vish.mk.