Roberto Díaz

@IgorBrigadir It seems that almost nobody has seen Oppenheimer.

I was one of the early testers of Mythos. I signed an NDA, so all I'm gonna say is: Wear sunscreen. If I could offer you only one tip for the future, sunscreen would be it. The long term benefits of sunscreen have been proved by scientists ...

@altryne Just issued a "ticket" to @AnthropicAI because of this. Never happened before.

My feed is showing me a bunch of folks who tapped out their whole usage limits on Mon/Tue. Is this your experience? Please comment, I want to understand how widespread this is

@ai I have no problem with more "freedom". The problem is that I can't take for granted that elites/governments will allow us to just exist. Much less let us enjoy material abundance. In any case...we'll soon see how this unfolds.

It’s 2026. You watch patiently as your delivery robot swerves to avoid the human body splayed out on the dirty, sunlight-dappled sidewalk to deliver the pizza you agreed to pay for in four installments with a $3 delivery fee + $4.99 service fee at 35.68% APR on Klarna.

Self-directed learning through AIs is an autodidact’s paradise.

@DaveShapi @arnauramio For the current state of things in the west, unemployable == expendable

I have revised my estimates for "future employable humans" For reference, my last work estimated around 20% to 25% total labor force participation rate. However, as I've refined my approaches and assumptions, that has been revised down to a LFPR of only 15%. That means that, in the future, I anticipate that less than 1 out of 6 working age adults will have meaningful employment. That may sound abysmal, but the solution is elegant.

Can't wait for fully automatized factories to be APIfied.

The more I create (not just software) with claude code + Opus 4.5, the more underhyped I see AI.

We are going to miss so much the problems we used to have until 2025.

@DaveShapi @MarceloLima Also, we need to have in mind that AI can learn virtually in no time every new discipline that could spawn after another one is automated. Prompt engineering is the silliest example.

You gotta run the numbers. There's not enough human attention to go around to create enough "more jobs" for that to really matter. Attention economy is very tail heavy. Experience economy is winner takes all. You're not wrong that an automation-heavy world creates a few more opportunities but the jobs that remain break down into two axis: 1. The attention/meaning/experience/authenticity/transformation axis 2. The statutory axis (jobs that are legally mandated) Neither will absorb most of the economy.

They're catching on.

Cost of opportunity will get in the middle at some point. For a small app I'm building that's gaining traction, I decided to create my own roadmap/feedback solution. Though it's working, I realized that the more time I'm investing in developing it, the less time I've dedicated to my app. Yeah, I got several agents working in parallel etc. But cost opportunity is real and, in case the project grows, the cost of mantaining that even with agentic support will outpace delegating it to a good product like @UserJotHQ

This is the way. Existing SaaS vendors are cooked.

@obie For now. I'm greatly surprised with the problem solving skills and creativity (not just for coding) that Opus 4.5 has.

Those 10,000 hours you spent becoming a great programmer aren’t useless at all. They’re what give you the two things that actually matter now: judgment and taste. Judgment is knowing when what the AI produced is correct, robust, and appropriate versus when it’s subtly wrong, brittle, or dangerous. Models can generate code. They can’t reliably evaluate it in real context. (Or at least for the forseeable future, they are really really slow at it.) Taste is knowing what to build, how it should feel to use, where to draw boundaries, and when something is “good enough” versus genuinely good. It’s the difference between shipping output and shipping a product. AI is commoditizing keystrokes, not expertise. The bottleneck has moved up the stack from typing code to deciding what’s worth shipping and standing behind. Your years weren’t spent learning syntax. They were spent training your instincts.

@big_duca Felt the same...gradually during the past 3 years. But there's no turning back. Adapt and look ahead, valuing how we got here.

I am not sure if other developers feel like this. But I feel kinda depressed. Like everyone else, I have been using Claude code (for a while, it’s not a recent thing lol). And it’s incredible. I have never found coding more fun. The stuff you can do and the speed you can do it at now. Is absolutely insane. And I’m using it to ship a lot. And solve customer problems faster. So all around it’s a win. But at the same time. The skill I spent 10,000s of hours getting good at. Programming. The thing I spent most of my life getting good at. Is becoming a full commodity extremely quickly. As much fun as it is. And as much as I like using the tools. There’s something disheartening about the thing you spent most of your life getting good at. Now being mostly useless.

Please and thank you

@dwarkesh_sp @pawtrammell Once labor is fully automated, at least 5-6 billion people become expendable to these oligarchs. They have no incentive to provide any form of welfare to these populations.

New blog post w @pawtrammell: Capital in the 22nd Century Where we argue that while Piketty was wrong about the past, he’s probably right about the future. Piketty argued that without strong redistribution of wealth, inequality will indefinitely increase. Historically, however, income inequality from capital accumulation has actually been self-correcting. Labor and capital are complements, so if you build up lots of capital, you’ll lower its returns and raise wages (since labor now becomes the bottleneck). But once AI/robotics fully substitute for labor, this correction mechanism breaks. For centuries, the share of GDP that goes to paying wages has been 2/3, and the share of GDP that’s been income from owning stuff has been 1/3. With full automation, capital’s share of GDP goes to 100% (since datacenters and solar panels and the robot factories that build all the above plus more robot factories are all “capital”). And inequality among capital holders will also skyrocket - in favor of larger and more sophisticated investors. A lot of AI wealth is being generated in private markets. You can’t get direct exposure to xAI from your 401k, but the Sultan of Oman can. A cheap house (the main form of wealth for many Americans) is a form of capital almost uniquely ill-suited to taking advantage of a leap in automation: it plays no part in the production, operation, or transportation of computers, robots, data, or energy. Also, international catch-up growth may end. Poor countries historically grew faster by combining their cheap labor with imported capital/know-how. Without labor as a bottleneck, their main value-add disappears. Inequality seems especially hard to justify in this world. So if we don’t want inequality to just keep increasing forever - with the descendants of the most patient and sophisticated of today’s AI investors controlling all the galaxies - what can we do? The obvious place to start is with Piketty’s headline recommendation: highly and progressively tax wealth. This might discourage saving, but it would no longer penalize those who have earned a lot by their hard work and creativity. The wealth - even the investment decisions - will be made by the robots, and they will work just as hard and smart however much we tax their owners. But taxing capital is pointless if people can just shift their future investment to lower tax countries. And since capital stocks could grow really fast (robots building robots and all that), pretty soon tax havens go from marginal outposts to the majority of global GDP. But how do you get global coordination on taxing capital, when the benefits to defecting are so high and so accessible? Full automation will probably lead to ever-increasing inequality. We don’t see an obvious solution to this problem. And we think it’s weird how little thought has gone into what to do about it. Many more thoughts from re-reading Piketty with our AGI hats on at the post in the link below.

@andrealbts @ImSh4yy @UserJotHQ Thank you!! I ended up...building my own feedback system for some reasons I've shared with S. But the product is getting solid traction, so I may have to his route again at some point 😅

@_rbart_ @ImSh4yy @UserJotHQ Congrats on the launch! I'd go with the app subdomain since that's where users will see the widget.

@ImSh4yy Hi man. I want to use @UserJotHQ for my just launched app brakinglab.com Should I add the app subdomain or the landing page one? I plan to put the widget inside the app so hence the doubt. Thank you, love the product!

Just created the fastest @iRacing series planner possible. @astrodotbuild performance is something else: brakinglab.com/en/iracing-pla…

@peer_rich Funnily, LLMs will create much more maintainable and clean code than us. By far.

i was just thinking how much i disliked refactoring and merge conflicts and AI is surprisingly good at it wanna swap a library? AI does that wanna split up your messy single file code into components? AI does that wanna fix all “any” in your typescript mess? AI does that

You probably aren't a simracer but, in case you are or are curious, you can use it 100% free here: brakinglab.com

It's crazy because, without AI, I could have easily spent a month or more using all my free time. Instead, I did it much better in about a week and a half. That's nuts.

🧵 Built BrakingLab during Christmas break: brakinglab.com A companion app for sim racers (mainly @iRacing and @LeMansUltimate) that helps them: - Perfect braking technique - Train laps based on telemetry - Schedule and prepare races for their series