idriss

I caught mine mining btc

Fox News host aged like shit in one year

have a really bad dev next to you? just hire the same dev a second time and tell them they are now full time reviewers.

not in a million years

yep, instant skip

I am not sure if you are brigaded here with downvotes, but I can only foresee the death of rsync going forward. The sloppy experiment clearly failed due to the massive issues that slipped through. He is doing it for free, I get it, he has the freedom to do what he wants but we can also jump ship to something with less features and no slop

good if the laws are approved by citizens (you don't get 40 years in prison for skipping tax vs free pass for pedos) & applied to EVERYONE (you skip tax you get 40 years, parasite twitter guy does that and he get praised of how smart he is

This has to be one of 3 things:

• Their bug/code base is really trivial
• They are not good at their job and missed important things
• The Claude code they are using is really different than the one I am using

Because I was experimenting with LLMs since gpt 3.5, still use them on and off because FOMO and they are still shit for anything that's not trivial

We have windsurf review bot (uses Claude under the hood I guess), it has 3 states: LGTM, crashed you should try later (never works even after 1000 retries), or the change is too big and refuses to review it.

I am nowhere close to be as smart as him or people like him, but I abandoned a project because of sloppy massive PRs and started refusing to put my name / approval on most PRs at work also because of the same thing, I am the tech person who stayed the most here and not gonna waste my time reviewing slop for others to take credit (just push it with Claude, dev XX will review it properly before it hits prod).

The decline started, but I will keep protecting my side in this company while I am here. I use LLMs occasionally (so I am not going against management trying to make us more productive) to type less, or extract smaller snippets, ... but the change is 100% what I had in mind and I hold myself 100% responsible for the output.

who needs calculator when you have next-token-predictor

What's the biggest code base you have worked with?