E.g. database optimisations a Senior Engineer might do, such as designing a database partition or creating a complex composite index. The problem? When Claude recommends more advanced solutions without a deep understanding it is easy to miss where the foot guns lie or if Claude got it outright wrong.
It's like being handed a chainsaw when you had an axe. Without good judgement, it's easy to cut down the wrong trees.
Think about what we did before if we didn't have another human around to ask and think together about a problem.
We searched for solutions or more info on Stack Overflow, Reddit , random blogs or HN even. The we tried to evaluate the pros and cons of each possible solution and then decide what to do.
Now we should use the LLM to get that info from the internet (be it from its lossy memorized or better fresh from its search tool). Then try to ask the LLM for pros and cons and follow the links it provided if you don't trust its "judgement".
This is an area where Claude Code is both valuable and dangerous. It can propose sweeping (correct) changes based on inconsistencies it finds within the codebase. The developer, in situations where nobody more senior is around to answer those design questions, is left making a judgement call based on vibes and what logic they can piece together about Claude's changes.
> I love to solve problems using Web Technologies. JavaScript/TypeScript (along with the foundation, that is, HTML and CSS)
"coding" is "dead" for Web Technologies and frontend roles (for those looking for inflated starting salaries at $200K+) because AI can do at least 98% of the work and can at best one shot this sort of work.
I see...
For whatever high percentage of engineers, having AI generate and edit code is now a large part of their day. That and reviewing code and testing take up more time.
Whether you want to call them engineers or not, producing custom software is much more accessible now.
There are a lot of consequences. For one thing, I think that this is going to reduce the market share of products like Salesforce and some other relatively high priced software that is often highly customized. There will be lots more open source competitors and many companies or departments generating custom software to replace it or parts of it.
Long answer [^1]:
> Betteridge's law of headlines is an adage that states:
> "Any headline that ends in a question mark can be answered by the word no."
[^1]: https://en.wikipedia.org/wiki/Betteridge%27s_law_of_headline...
You could be an amazing player, but everyone will point out the cheat codes are on. The last refuge will be deeply niche programming or areas not well represented in, or not generalizable from, the vast training corpus.
digitaltrees•1w ago
yomismoaqui•1w ago
Seriously though, now I think more about architecture and testing than before. Also I end the day with less foggy head than when I hand coded.