lazopm

I wonder how much of this is due to politics. I always thought it wouldn't matter much to most people but earlier this year I tried to convince my wife to get a model Y for the family and there was no convincing her. The interesting thing is she doesn't even care about politics but just didn't want to buy a car that is associated to a certain political view. I guess that happens when you take such a hard stance, you alienate one entire side and the people in the middle that wanna stay out of it.

Is PDF the new Flash? It just does too much for its own good, it has become a huge attack vector

The way I see it you should calibrate the way you work with LLMs based on how confident you are on that specific area, and if it's your responsibility to own/understand it. Here's how it feels for me:

* Learning stage: 0.5x - 1x. I change my system prompt to teacher mode, taking the productivity hit for actually learning the system/tool pays off dividends later. I change my system prompt to "teacher mode" and slowly loosen it as I get more confident.

* Working-knowledge: 2x - 3x. Once I am ramped up enough I feel like I can get a decent productivity boost. Most of the time is spent at the planning stage. This is my mode for areas I don't really own or care about, just need to get work done.

* Mastered: 10x+ I have been doing web front end for 12+ years, I can quickly review plan/implementations and for my initial prompt I already know most of what I want built.

1x == my speed before AI

Different things work for different people, personally white boarding is my favorite. I think the best would be to offer the candidate options. Let them come in for the culture fit portion of the interview, at the end present them a problem and let them choose if they prefer white boarding, pair programming, or take home and submit by the end of the day.

I think you can just iterate over the text nodes and see if you stumble upon every letter you're looking for in the right order, it would work for any kind of randomly added text.

It's a way to preemptively send assets to the client before they request them.