Second Thoughts

> I am not a coder (much), but I have to wonder if my experience is common in the coding world? It is, yes. Surely someone will come and tell you it doesn’t happen to them, but all that tells you is that it ostensibly isn’t universal, but still common enough you’ll find no end of complaints. > Maybe there are ways to avoid this (and others) pitfall with prompting? Prompting can’t help you with things not in the training set. For many languages, all LLMs absolutely suck. Even for simple CLI tools, telling an LLM you are on macOS or using the BSD version may not be enough to get them to stop giving you the GNU flags. Furthermore, the rsync change in macOS is fairly recent so there’s even fewer data online on it. https://derflounder.wordpress.com/2025/04/06/rsync-replaced-with-openr… > As it is, I do not see at all how LLMs could really save time on programming tasks without also costing more time dealing with its quirks. And that’s the best case scenario. It also happens that people blindly commit LLM code and introduce bugs and security flaw they cannot understand or fix. https://secondthoughts.ai/p/ai-coding-slowdown https://arxiv.org/abs/2211.03622

For people who prefer reading to watching videos, I wrote a detailed account of my process for solving one of last year's IMO problems, along with thoughts on how this relates to AI: https://secondthoughts.ai/p/solving-math-olympiad-problems

Theres an interesting recent study showing how poorly developers measure their own productivity: https://secondthoughts.ai/p/ai-coding-slowdown HN discussion: https://news.ycombinator.com/item?id=44526912

It's funny how that works in your mind. There comes a snake oil salesmen and says that LLMs are the bees'knees and you just take as a given without any critical thought. But when people actually have proof that LLMs are nothing special and even reduce productivity (by 19% in experienced developers): https://secondthoughts.ai/p/ai-coding-slowdown you are screaming to the heavens for proof. I am the last to stop the AI fanbase from outsourcing their critical thinking and some of their wallet to a fancy autocomplete, that means more work for the rest of us that actually do the thinking. LLM vibe-"engineers" will be fleeced $$$ when Anthropic actually starts wanting to be profitable: https://news.ycombinator.com/item?id=44598254#44602695 Their skills deteriorate, and so does their critical thinking: https://www.mdpi.com/2075-4698/15/1/6 And I just can't stop giggling imagining their workday when Claude is down. LGTM is not a sustainable career choice, people.