← Back to context

Comment by neya

19 days ago

As an engineer, I can never actually let a system write code on behalf of me with the level of complacency I've accumulated over the years. I always have opinionated design decisions, variable naming practices. It's memorable, relatable, repeatable across N projects. Sure, you can argue that you can feed all this into the context, but I've found most models to hallucinate and make things unnecessarily opaque and complex. And then, I eventually have to spend time cleaning up all that mess. OP claims they can tell the model over the phone what to do and it does it. Good for OP, but I've never personally had that level of success with my own product development workflow. It sounds too good to be true if this level of autonomy is even possible today without the AI fucking something up.

I've started collecting all my opinionated practices in a file I call ENGINEERING_PRINCIPLES.md, which I share across all my projects and reference in every CLAUDE.md. It contains variable naming practices, rules of thumb for writing good code comments, guidelines for unit & integration tests, general architectural principles (avoid global state & push side effects to the boundary aka "functional core, imperative shell", avoid Java enterprise-style OOP hell, etc.), …

So far, the code Claude has generated looks fairly decent and stylistically not too different from what I would write myself.

> I always have opinionated design decisions, variable naming practices

That's the source of your difficulty. Research wu-wei.