nemo44x

I live this site - down votes even though it’s 100% true. There are TONS of old recordings that don’t even have masters anymore and the digital versions are ripped from a vinyl copy. There’s entire communities around this.

There’s a lot of articles about it. It costs them $500+ for heavy users. They do this to capture market share and also to train their agent loops with human reinforcement learning.

https://ezzekielnjuguna.medium.com/why-anthropic-is-practica...

They lose money at $200/month in most cases. Again, the old rules still apply. You are the product.

Sentimental sure but there’s also a lot of music that was never properly remastered for digital so the vinyl distribution does in fact sound better.

Has anyone started a project to replace Linux yet?

We need filters. I’m not agreeing with Thiel but pointing out to parent why he is listened to but a guy sitting in his own piss is not.

Correct. We call that “credibility”. People can earn credibility in this world through their accomplishments. If your life accomplishment is sitting in your own piss you lack credibility.

Yeah I don’t think yet. But keep generating and tooling will arrive that can better connect it. We have the nodes (markdown files) but not the edges of the graph.

This is actually a good thing, or could be, for you and here’s why.

Today no one gets a call back when they apply for a role they’re 100% qualified for. It’s because the recruiting system broke when LLMs became widely available. You can dump a job description in and generate a fake CV and cover letter with no effort.

People apply for every job. A job listing that would get a few hundred good applicants get thousands of perfect ones now. It’s all noise. And it’s getting worse.

Not that this is the only solution but an AI screening call would give qualified applicants a chance to distinguish themselves and get to the human phase. It could incentivize people to not apply for everything.

It’s becoming a bigger and bigger problem actually. AI can create a perfect CV and cover letter and people apply for everything. It’s hard to get a callback because you’re just 1 datapoint. So recruiters have to call you first or use recommendations.

Provisional ballots work fine. Statistical inference works too.

At the very least teams will communicate with each other much better. So much of the tedium of office work is able to be automated so people can spend more time solving problems instead.

But the communication will massively improve. More artifacts being generated of progress and needs and AI can link related things around an organization rapidly and accurately. Workflows will massively improve. A living graph of an entire organization will come to life.

I think more productivity gains will come from this automation than anything. People will look back at all the drudgery workers did.

Paper ballots work just fine. Why are we using tools for scale (computers) when voting is an incredibly small and finite domain. Just total waste of tax dollars and over engineered solution to a simple problem.

How do people not understand this? LLMs are goal machines. You need to give them the specific goal if you want good results and continue to reenforce it. So of course this means speccing and design work.

People are so enamored with how fast the 20% part is now and yes it’s amazing. But the 80% part by time (designing, testing, reviewing, refactoring, repairing) still exists if you want coherent systems of non-trivial complexity.

All the old rules still apply.

How many jobs were had or not due to the candidates ability to implement his algorithms?

They are, but the dogs got it pretty good now. Good shows for good boys.

So their business model is to deliver me buggy code and then charge me to fix it?

Imagine owning a money printing machine and then breaking it because you want it to print more money more quickly.

Supervisor agent frameworks are going to be a big industry soon. You simply can’t have agents executing commands without a trusted supervisory layer examining and certifying actions.

All the issues we get from AI today (hallucinations, goal shift, context decay, etc) get amplified unbelievably fast once you begin scaling agents out due to cascading. The risk being you go to bed and when you wake up your entire infrastructure is gone lol.

You need to be sociopathic aggressive right now. We are in a few month window here where execs are very confused even if they parrot their group chat talking points to sound smart and ahead of it. They are all behind and they know it.

This isn’t comfortable but now is the time to ship fast and hard. To overstep boundaries and be the person getting attention. In a few months everyone will be so you need to do this now.

Just don’t. Don’t limit yourself. Ask for forgiveness.

It’s very uncomfortable. But yes, need to figure out how you differentiate and create extreme value with AI.

We went from “how can AI be useful to me” to “how can I be useful to AI”. Or headed that way.

The entire system and the agent loop allows for more complex goal resolution. The LLM models language (obviously) and language is goal oriented so it models goal oriented language. It’s an emergent feature of the system.

Work in smaller parts then. You should have a mental model of what the code is doing. If the LLM is generating too much you’re being too broad. Break the problem down. Solve smaller problems.

All the old techniques and concepts still apply.

LLMs are literally goal machines. It’s all they do. So it’s important that you input specific goals for them to work towards. It’s also why logically you want to break the problem into many small problems with concrete goals.

Database migrations are hard and inductive and often fail in some aspect. Why would you want to spend time doing them when you can spend time building the important thing after the migration is done.

Secondly - AI helps with happy path tasks for a migration. But most database migrations are complex beyond what an LLM can just spit out. There is so much context outside the observable parts of the database AI has access to. So I don’t think you have to worry about vibe coding eating the entire migration project.

Yeah building acceptance criteria first is the way. An LLM is a goal machine. It uses probability over and over to advance towards the goal(s). That’s all it is and wants to do. So giving it well defined and granular goals and guardrails will get the best results.

I’m implying that going anywhere in the Middle East (or a good deal of the world) has far more risk than going to the USA.

It’s why I said not those exact things. But already today when developing with agents you get best results from selecting it as much as possible so it can then logically break it into smaller contexts and iterate.

It’s not just dropping in case and getting a functional product. But more up front design seems to work best. Very explicit design.

I have some Euro friends that went to Dubai instead because the USA is “too dangerous” right now. I wish them safe passage back, but yeesh talk about irrational perspective.