@Jouhatsu_ai The tweet is just a click bait (like most of the tweets about AI these days). I would be very disappointed if I paid $500 for this presentation — worth no more than 1 hr of your time on 2x speed. It is highly theoretical intro into building agents, not a workshop.
I can build ~2 (features | refactoring | brainstormings) with Claude per day on Pro plan with 6hrs token reset. Sonet+Opus Xcode setup gives a bit more extra tokens than just Opus only. canveman plugin only saves my time to read Claude response. While waiting - I code myself.
I wonder, why do I need to teach Claude that before removing the header import, it needs examine its content, as there might be C-funcs that the class calls. And it is just an example, but the misses on this level are really disappointing.
@Clara_Gold Matches my experience almost exactly. Lived in SF 5yrs. Made no real friends, while trying hard. Girls I really liked were always up in the sky having more options they can possibly consider. No sense of community. Only thing people care when they see you is what you do for work.
There's a new kind of coding I call "vibe coding", where you fully give in to the vibes, embrace exponentials, and forget that the code even exists. It's possible because the LLMs (e.g. Cursor Composer w Sonnet) are getting too good. Also I just talk to Composer with SuperWhisper so I barely even touch the keyboard. I ask for the dumbest things like "decrease the padding on the sidebar by half" because I'm too lazy to find it. I "Accept All" always, I don't read the diffs anymore. When I get error messages I just copy paste them in with no comment, usually that fixes it. The code grows beyond my usual comprehension, I'd have to really read through it for a while. Sometimes the LLMs can't fix a bug so I just work around it or ask for random changes until it goes away. It's not too bad for throwaway weekend projects, but still quite amusing. I'm building a project or webapp, but it's not really coding - I just see stuff, say stuff, run stuff, and copy paste stuff, and it mostly works.
This talk actually helps Swift engineers to understand where they are in Swift language expertise. It also helps to evaluate the other engineers’ level of expertise, if you need to do so.
"Build a balance bike, not training wheels." Doug Gregor's @swiftcraftuk talk describes Swift's progressive disclosure: start with the simple stuff, reach for the advanced stuff when you need it, and never have to throw out what you've learned.
https://t.co/WGcCU0vdsk
Airbnb CEO, Brian Chesky, on the most effective way to eliminate meetings:
"The reason there's too many meetings in a company isn't because they don't have meeting-no-meeting Wednesdays. It's because they have too many people."
He advocates for a small, elite workforce over bloated teams:
"We want a small, lean, elite, highly skilled team, not a team of kind of mid-level battalion type people. And the reason why is every person brings a communication tax."
He explains how mediocre hiring compounds into dysfunction:
"The old saying: A players hire A players. B players hire C players. I would like to amend it: B players hire lots of C players. Because those are the kind of people that like building empires."
And those C players can't do the work alone, so they hire more people, who go in different directions, creating even more meetings and admin tax.
His solution at Airbnb was radical—eliminate management layers entirely:
"I got rid of layers of management. I went back to functional. You can only manage the function if you're an expert. You don't manage people. You manage people through the work."
He credits this philosophy to Jony Ive, who believed most design leaders get it backwards. Ive's approach was to focus on the work itself and build a team that designs together.
Elite teams don't need elaborate management structures — they need leaders who manage through the work itself.