I've tried both, and I'm still not sure. Claude Code steers more towards a hands-off, vibe coding approach, which I often regret later. With Copilot I'm more involved, which feels less 'magical' and takes me more time, but generally does not end in misery.
Pretty much. If all you do is prompt in Claude why the hell you need to memorize how to solve leetcode problems? And usually the companies that push hard for leetcode are the one who also force to use AI to code.
I feel like building your own projects(for potential commercial success or for just a practice) is better preparation for an interview.
Meta by the way is changing the interview process and it's not leetcode anymore but AI assisted problem solution.
You've been downvoted, but this is exactly how I feel as well. There is an element of denial amongst the HN crowd, commonly saying stuff like "Well I tried Claude code and it produced garbage". Any task with a tight "write->test->repeat" loop is going to get AI-trained into oblivion, and we've only 4 years into this LLM disruption.
Personally, I'm training up on: infrastructure, systems administration, security and software-architecture - because these are harder to train on given a longer "write->test->repeat" cycle, although I'm not in denial that they too will be disrupted.
Writing code for any problem - especially Leet-code style problems - is going to be solved by AI eventually. Don't be left behind.
I hosted this Hack Day up in Seattle at the GitHub office and this project was one of 10 from an event that brought in nearly 60 project submissions! @toobulkeh is an awesome builder and was a pleasure to see this project come to life in a few short hours.
Actually just jobless, but I was doing side projects here and there
Retirement gets boring fast —- and you lose connections to the rhythm of society fast.
For the vast majority of humans, an idle mind is depressing and destructive
reply