A very very senior game developer recently told me how he is changing specialization because AI is much better than him at what he does already.
I asked him if it was just the speed and mass of code, to which he replied that Claude 5 Fable/Mythos is just so advanced, it comes up with solutions he would never have thought of, not even requiring good prompts anymore.
He further said in some areas like Unity it's not perfect yet, but improving drastically.
Honestly this is pretty different to what I read in many articles regarding AI code.
They mentioned errors, bad performance & such.
However if it is already used in production and he as a professional evaluating it, is saying it's better than him...
That same day I also talked to somebody that got hired to train an LLM. His interview was with that companies chatbot, that offered him over 120$ per hour, a roughly 500% increase from his previous job š¤£
I dunno, maybe they aren't that senior then? Just because it came up with a different solution doesn't mean it's the right or appropriate one. I've seen LLMs output very clever code for the wrong situations entirely.
People have also been really impressed by simply functional code, but the architecture it was written for was completely incorrect. It barely took the existing code and structure into account. It was also my job to review AI code written by models just weeks ago. It had been prompted and reviewed by senior devs and I couldn't stop finding extremely questionable code e.g hundreds of database requests in a function that was called in a loop. They had beefy dev machines which were OK with that but the production environment had the power of 200ā¬ mini PC and limited RAM.
With code like that solutions like "let's scale vertically and horizontally" actually become costly. The cloud bill has been stradly increasing and jumped 30% in one quarter. I'm afraid of what will happen if they let the AI loose on the cloud and give it full access via Terraform. Their bill will probably be integer multiples not just floats.
save