rafaelmn

If the option is 25k for a car I "can't repair" in 5 years or 50k for a car I can - I'll take the 25k one, if I can get the warranty to cover those 5 years, all else being equal.

Everything is "repairable" it's just a matter of is it worth repairing. Maybe rebuilding the car every 5-10 years isn't that bad of a model.

>Consumers want protection from cars they can't afford to fix.

Said who ? If it's under warranty for long enough and cheap enough then I don't really care even if I have to fully replace it in 5 years and sell for scrap. Without tariffs that's what the price would be.

But if that 20% is not the bottleneck then at most your market is replacing the labor required to get that 20% to the next phase, plus you will probably not be able to capture full value here because of lab/model competition.

Compared to Air 2 :D ?

Reading boost libraries was like reading brainfuck. In fact most of std lib source felt that way too. I never reflected on this since I stopped doing C++ 15 years ago - but yeah I have not worked with a language that's worse to read since. Maybe macro heavy lisp ?

That's why OpenAI hard pivoted from consumer tech into enterprise ?

The only AI that makes money is enterprise. Being open weights/self-hostable is a huge win before the pricing even enters the conversation, but pricing is also adding a ceiling on what US labs get to charge. And stuff like powering other software also works better with fine-tunes, hence legal models, etc.

Agents are pretty shit at making UX decisions unless you're heavily in the training distribution

Depends - if the model asks for permission and you approve - yes. If it's auto mode or something like that then it's probably on authors despite what they claim in ToS/EULA

Realistically once models are good enough at this most of us are out of a job.

Embedding@Work

TBH I'm seeing the opposite, I have a legacy codebase where the people originally writing it were incompetent and inexperienced, but the project took off. It's scaling up but the foundation is shit and has to be gradually rebuilt. Every AI slop commit is still better than the underlying crap.

And this is the pattern I've seen on most (semi)successful projects I've worked on in the past - I'd say correlation between financial success and code quality is 0 (up to a point where the whole thing doesn't fall apart). Once scale (both in load and in code size/features) starts mattering you're stuck building on a foundation of shit. LLMs are very good at identifying and cleaning up said shit layers, as long as you're steering them towards a desireable outcome.

> Your code used to be a masterpiece

Do you even see anything underneath that rose tint ?

A smarter agent can write a better compiler and leave the intent in densely packed/verifiable language structure that can encode propositions low level code cannot.

This is such a bad take given the current LLM agents

High level languages :

  - compress low level patterns into high level abstractions which save context/token
  - express higher level meaning through things like names
  - express and enforce constraints through type system

LLMs are context constrained, do really well with tools that help them iterate towards a correct solution (type systems) and work on human language (although they are compressing it to be more token efficient).

Why would you take something compilers are really good at (translating high level concepts into low level hardware instructions) and bake that into LLMs

If anything LLM programming language will be something like high level IR that's token optimized, but honestly the AI labs are so lazy (or bad) at fundamental engineering (look at their sandboxing solutions LOL), they are going to keep pumping training on shell/python/rust RL environments.

No, not nearly the same living standard.

Entirely different generation

You do realize that St Petersburg dot on the map is Russia ?

How does the model get context to decide ABCD ?

Look into beads/dolt then - it does this pretty much with a cli - Jira for agents :)

That's the reason I'm excited about the orchestrator because it would let me build my ideal UX on top.

If you move away from the idea that agents are chat streams and treat them as processes/actors then you can start letting agents represent themselves/build their own interfaces.

And if you build enough introspection into the protocol because everything is message based you can have other agents build interfaces for them.

So like either standard GUI, or a voice assistant talking to you and delegating to agents, etc.

It's not an idea it's a way of thinking about agent systems - basically an agent OS.

I'm actually taking the time to hash out the details to do this as a side experiment project, but basically what I read this project as showing is that if you leverage asynchrony you can let agents be more efficient - and you get this "for free" if you model your "agents" as actors that interact with the system through message passing. Then all the system operations become messages to different actors, need to read a file => message the fs actor => get reply from FS actor as a message when it's done. You'd probably need some out of mailbox ways to share blob resources and sockets for realtime (audio), but for the most part simple message passing should handle most of what LLM agents do.

Actors can have identities and roles for RBAC, etc. you - cross agent communication is the same as sending any other message to a actors.

Not to mention that actors can be on your device, another device, etc. whatever the router can resolve - it's transparent to the agents.

Honestly the fact that we're still modeling agent harnesses like chat and strapping them into shell sessions instead of building async actor systems with sandboxed OS functionality access actors is bananas to me. So much easy wins can be had just by building on right abstractions... Hopefully will get enough time to play with this idea soon on my own.

Lack of way to capture value from intellectual labor is considered a market failure that leads to suboptimal market results for consumers, but with AI that argument becomes very weak.

Sorry but the point isn't to create artificial scarcity just so intellectual labor is well off, that's a negative side for the consumer that was considered necessary tradeoff. Market economy should be about providing the most value to the consumer.

Disclaimer - I was never a fan of IP laws despite them working in my favor, with AI I can see them finally being abolished.

Copyright is artificial scarcity rationalized by arguing that producing novel intellectual work is valuable, but requires substantial effort that can't be recouped, so we have to incentivize it somehow.

LLMs and AI are changing that proposition substantially - human effort involved in producing copyrightable content is getting reduced constantly to the point that if we abolish copyright entirely we'll still have more content than we could ever hope for.

AI/robotics eliminating scarcity of physical goods sounds very far fetched but in the intellectual space it looks very very plausible in the near future - so it could be time to abolish IP laws soon, especially if AI manages to advance enough in R&D and research space.

You don't have to have a JIT to have a JS VM ?

There's also a recency bias, where you ignore how valuable some decision was in the past and ignore the fact that you maybe wouldn't even get to the place you are if "you made a different decision in the past".

Debugger MCP better

Meh I just replaced a battery in my 14 pro (and the charging port) for ~200 EUR. I didn't want to upgrade yet and I would not even be considering upgrading if the device had USB-C

How exactly ?

We heard this since Uber days and all it did was break the old taxi systems and brought taxi service to a wider audience. As a consumer I benefited immensely from Uber.

Likewise for AirBnB and hotels.

Not saying that these don't have secondary effects on the society - but as a consumer they were and arguably still are pretty amazing.

You can easily setup codex rules to auto approve local stuff but gate external effects like push/jira write/curl - works better for me than full yolo mode - depends if it's a solo project or working with a team.