@henriquebastos@gawry@wescld@sseraphini@sebastianconcpt Eu bem que queria comprar essa briga, mas as regras de compliance são muito rígidas
Até dá pra usar OpenCode e Pi, porque eles integram com o OAuth do GHCP de forma nativa, mas nada muito fora disso.
É muito engraçado pensar que na década passada isso surgiu como piada, e parece que um roteirista do candidato era fã do @DanielSFurlan
É o que releitura que chama isso né?
Obrigado @amadafoca por ter nos proporcionado isso uns 12 anos atras
Amongst my friends, Spotify is the lowest quality consumer app we still pay for. It certainly hasnt gotten noticeably better in the last couple years (arguably worse). So, this is not the positive look Ant and Spotify are spinning here.
Bigger picture, this is the problem with a lot of AI reporting. It reports completely meaningless metrics like deploys per day or LoC. Why don’t we start reporting consumer satisfaction reports? Actually end state research results.
All the no nuance AI people always come out and think that this is anti AI. Again, I think AI is great and Claude is great. But this is bad marketing and makes both look like clowns.
@gabrielmontsms@ayubio o 2o, 4.228.31.150, os response headers mostravam que realmente era do github:
...
< location: https://t.co/FndPEfOrpe
location: https://t.co/FndPEfOrpe
< strict-transport-security: max-age=31536000; includeSubDomains; preload
...
* Connection #0 to host 4.228.31.150 left intact
@gabrielmontsms@ayubio o 1o, esse mesmo 4.228.31.149 e dava timeout:
curl -4Iv --connect-timeout 5 --insecure https://4.228.31.149
* Trying 4.228.31.149:443...
* ipv4 connect timeout after 5000ms, move on!
* Failed to connect to 4.228.31.149 port 443 after 5004 ms: Timeout was reached
Told Cursor to build me a Slack replica.
It spawned 1500 AI agents.
Around half of them immediately became product managers.
They spent the whole day pinging the other half:
"any update?"
"just checking in"
"circling back"
"friendly reminder"
By the end of the week, nobody had built anything substantial, but somehow we burned $25k in LLM tokens.
I think it may have been over-trained on corporate data.
@henriquebastos@mattpocockuk I used it for a deep dive on an app I’m building for personal use, thought it’d be easy.
It hit me with 76 questions and uncovered edge cases I didn’t even know existed 🤣
Really good
@plainionist Learn the basics. Just as before. Learn them very well. Write assembler, C, Java, Ruby. Learn algorithms and data structures. Read the old classics.
And then start using agents.
Novices with power tools tend to lose fingers.
Here’s @dhh (Creator of Ruby on Rails & Omarchy, cofounder of 37 Signals) on why using AI agents does not feel like being a Project Manager:
“When I was on that Lex interview last summer, I was saying: I don't want to be a project manager for agents because I had the mental model of a project manager of humans, and that's not what I enjoy. I don't want to be that far away from the production. I want to be in the mix. I want to have my hands in the code.
What I failed to realise at the time was that running a bunch of agents feels less like being a project manager for agents and more like stepping into this super mech suit, where suddenly I don't just have two arms, I have 12. I can now look at seven screens at the same time, running five keyboards. I'm still the one doing it, even if I'm not typing ‘this’ as a keyword in a program.
I have been hyper accelerated as a programmer. It's a different kind of programmer, but it still has the same affinity to aesthetics, at least when I'm producing Ruby code, and I'm able to combine that while being vastly more productive on a bunch of things.”
After years of keeping ideas and opinions only in my head (not that anyone asked), I finally decided to start writing publicly.
Software engineering, systems, infra, and whatever else my ADHD brain decides to obsess over.