Rodrigo Juliani

Software development, tech, building products and psychology. My Github profile sucks, so don't even bother.

[Rodrigo Juliani @dragonautdev]

Free speech may offend you, but it gives you the chance to object, discuss, dislike and express your discomfort. The alternative doesn't.

[Rodrigo Juliani @dragonautdev]

@OrsiYamandu Electricidad barata y Uruguay no son parte de la misma oración desde hace añares. ¿Estás bien?

[Rodrigo Juliani @dragonautdev]

@GBica1987 O sea... Totalmente posible que los números del canal estén inflados por medios artificiales... Pero en tu captura de muestran visualizaciones por hora. Mi auto tiene 67.000 km, y eso que hay veces que ando a 30km/h 🤷‍♂️

[Rodrigo Juliani @dragonautdev]

@atmoio If you had to clarify so much about the video, it's clear to me that those claiming things about LLM's greatness and the achievement of AGI have in fact been so far away from intelligence, it's even harder to believe they'd be able to recognize it. Kinda proves your point btw

[Rodrigo Juliani @dragonautdev]

@vimtor This is how I imagine playing dragonforce full 5 instruments with my laptop

[Rodrigo Juliani @dragonautdev]

@larrykim What's stopping you from automating yourself out of the job and have your company become an X^10000 company. 0 humans, X^10000 lines of code, 29/7 productivity. If you're not doing this, you're behind already

[Rodrigo Juliani @dragonautdev]

@theo Well yes I still do, and while some awesome alternatives/complementary tools have appeared, we're still a bit far from the point where I can trust those alternatives 👉 https://t.co/G4FfzZ6fzI

[Rodrigo Juliani @dragonautdev]

@adiix_official I'm sure if the github CEO is reading this... he traveled in time 🤣🤣

[Mitchell Hashimoto @mitchellh]

I strongly believe there are entire companies right now under heavy AI psychosis and its impossible to have rational conversations about it with them. I can't name any specific people because they include personal friends I deeply respect, but I worry about how this plays out.

I lived through the great MTBF vs MTTR (mean-time-between-failure vs. mean-time-to-recovery) reckoning of infrastructure during the transition to cloud and cloud automation. All those arguments are rearing their ugly heads again but now its... the whole software development industry (maybe the whole world, really).

It's frightening, because the psychosis folks operate under an almost absolute "MTTR is all you need" mentality: "its fine to ship bugs because the agents will fix them so quickly and at a scale humans can't do!" We learned in infrastructure that MTTR is great but you can't yeet resilient systems entirely.

The main issue is I don't even know how to bring this up to people I know personally, because bringing this topic up leads to immediately dismissals like "no no, it has full test coverage" or "bug reports are going down" or something, which just don't paint the whole picture.

We already learned this lesson once in infrastructure: you can automate yourself into a very resilient catastrophe machine. Systems can appear healthy by local metrics while globally becoming incomprehensible. Bug reports can go down while latent risk explodes. Test coverage can rise while semantic understanding falls. Changes happens so fast that nobody notices the underlying architecture decaying.

I worry.

[Rodrigo Juliani @dragonautdev]

@vimtor That and the insecurities it raises about one's own level on a language when it messes up half the words 😭

[Rodrigo Juliani @dragonautdev]

@PhDenLogica A ver, que esperamos de los "profesionales" que trabajan para el estado? Toda la gente uruguaya que conozco y es mínimamente decente en desarrollo de software trabaja en el ámbito privado y en general internacional. Lo que queda entonces es... Esto 🤷‍♂️

[Rodrigo Juliani @dragonautdev]

@vimtor If you have the chance, use depot or similar. Avoid gh actions as much as you can.
If you want to invest a bit more time in exchange for more reliability and control, it's possible to do your own thing inside a vps connected to github webhooks.

[Rodrigo Juliani @dragonautdev]

@vimtor @SST_dev You are doing an awesome work man!

Out of curiosity, is the Router component also coming to cloudflare anytime soon?
Thinking about it... Is it even possible to do it?

[Rodrigo Juliani @dragonautdev]

@shadcn Always. My users are (sadly) less technical and more old school, which means they expect to see something's clickable with the pointer icon.
I understand the design principles behind not doing it by default, but I'm not the user, not even close.

[Rodrigo Juliani @dragonautdev]

@PhDenLogica "12 metros cuadrados bien aprovechados " dice el que tiene el water de almohada

[Rodrigo Juliani @dragonautdev]

@vimtor You know how bad that sounds right? 😅

[Rodrigo Juliani @dragonautdev]

@theo I agree with you, and I've been saying the same for as much time as you or more.
But there's the hype crowd, the marketing crowd who absolutely think they've found an edge to get audience and sell a cheap-ass tutorial. Sadly CEOs (such as mine) are their target.
Wish me luck

[Rodrigo Juliani @dragonautdev]

@thdxr Consider a majority of users of openclaw can't really tell if the model is doing good or bad, specially around tasks that are deep in the OS or dev-related. Then it becomes more important for them to compensate their own shortcomings with the model. 🤖

[Rodrigo Juliani @dragonautdev]

@thekitze I would say most real use cases are for things that run on schedule or because of an external event.
I'm not saying it's impossible to do with opencode/cc/codex, but feels like openclaw is more built for that... And then you have it invoke opencode for the actual task

[Rodrigo Juliani @dragonautdev]

@nuxt_js I just had AI replace the entire react's codebase in one prompt. I don't know what it did, but was like 1 command and I had a whole copy of reacts codebase in my computer. Ready to ship for my 1T MRR solo agent-ruled company