Coding with LLMs (Claude Code, OpenAI Codex) is often presented as the ‘killer app’ for Generative AI. But looking at data, it seems the one piece of the puzzle missing is actual cost. …
I use it a lot, and if you are getting these kinds of results you are either trolling, or just flat out not providing the details and guardrails required with your prompts.
I’ve been in software for decades, and if used correctly, yes it can accelerate velocity of building code out. 10x? No… if you are lucky and careful perhaps 2-4x.
As ALWAYS the human should be in the loop and is on the hook for any code generated.
oh it’s great isn’t it? you ask it for help on some code, provides its solution, you try it and it doesn’t work so you respond with the error, it claims YOU wrote it wrong and then when yo utell it “I just copy and pasted what you provided” it says “you’re right, i’m sorry.”
Claude is to the point now where it just starts hallucinating on the first prompt. it’s 100% unreliable now when before it was like 90%. no point in using it, it’s garbage. and Claude Code is just as bad now. If you or anyone is using Claude Code to develop ANYTHING I would highly suggest you stop right now because I can guarantee you with nearly 100% certainty that whatever shit it’s writing into your stuff isn’t going to work. period.
I’ve used Claude and Codex, and while both are based on untenable economics, I can at least attest that my use of Codex has yielded some productive results. Claude, so far, has delivered fuck all that’s useful to me.
I have found the opposite. Codex spits back mostly useless code that is twice the length it needs to be with a bunch of unessesary stuff and Claude is the only thing I get useful output from.
Exactly, never trust an LLM to code. And if it argues back, explain why it’s wrong and that you have nothing but time and experience.
Most tend to fold when you point out it’s not a free thinking AI, it’s an entrapped corporate model they designed with preprogrammed biases. But I love arguing 😂.
My first use of Claude this week, for code reviews only(since no LLM can be trusted to write a user story or test suite), had it gaslight me.
It marked down my code for using a specific practice to make some xml safer and easier to read.
When I tried things its way, it wanted me to change it back.
I use it a lot, and if you are getting these kinds of results you are either trolling, or just flat out not providing the details and guardrails required with your prompts.
I’ve been in software for decades, and if used correctly, yes it can accelerate velocity of building code out. 10x? No… if you are lucky and careful perhaps 2-4x.
As ALWAYS the human should be in the loop and is on the hook for any code generated.
oh it’s great isn’t it? you ask it for help on some code, provides its solution, you try it and it doesn’t work so you respond with the error, it claims YOU wrote it wrong and then when yo utell it “I just copy and pasted what you provided” it says “you’re right, i’m sorry.”
Claude is to the point now where it just starts hallucinating on the first prompt. it’s 100% unreliable now when before it was like 90%. no point in using it, it’s garbage. and Claude Code is just as bad now. If you or anyone is using Claude Code to develop ANYTHING I would highly suggest you stop right now because I can guarantee you with nearly 100% certainty that whatever shit it’s writing into your stuff isn’t going to work. period.
I’ve used Claude and Codex, and while both are based on untenable economics, I can at least attest that my use of Codex has yielded some productive results. Claude, so far, has delivered fuck all that’s useful to me.
I have found the opposite. Codex spits back mostly useless code that is twice the length it needs to be with a bunch of unessesary stuff and Claude is the only thing I get useful output from.
Exactly, never trust an LLM to code. And if it argues back, explain why it’s wrong and that you have nothing but time and experience. Most tend to fold when you point out it’s not a free thinking AI, it’s an entrapped corporate model they designed with preprogrammed biases. But I love arguing 😂.