To think companies worry about protecting the talent supply chain is to put your fingers in your ears and ignore your eyes for the past 5-10 years. We were already in a crisis of seniority where every single role was “senior only” and AI is only going to increase that.

I actually think the opposite will happen. Suddenly, smart AI-enabled juniors can easily match the productivity of traditional (or conscientious) seniors, so why hire seniors at all?

If you are an exec, you can now fire most of your expensive seniors and replace them with kids, for immediate cash savings. Yeah, the quality of your product might suffer a bit, bugs will increase, but bugs don't show up on the balance sheet and it will be next year's problem anyway, when you'll have already gone to another company after boasting huge savings for 3 quarters in a row.

> obj-c runtime dynamism is fascinating. it’s like, “what if we make C as dynamic as Ruby”.

Well thats because objc and ruby are cousins. Both are sort of the only two smalltalk based languages out there

> Well thats because objc and ruby are cousins. Both are sort of the only two smalltalk based languages out there

I'm sure you can trace connections, at least in ideas, but I think Ruby is way more Perl-based than Smalltalk-based.

Im talking the fundamental language framework. 'Everything is an object' and method calls are actually message passing are the two reasons that objc and ruby are actually smalltalks.

I’ve tried to explain this before and unless you’re steeped in late binding, encapsulation, and message passing, the details are lost on most people (it seems including modern language designers).

For the GP, in most languages the dot or arrow operator is field access. If that field is a function reference, parenthesis are used to invoke it.

From outside of the object, neither Ruby or Objective-C allow direct access to object fields or functions. The dot operator sends the object a message that be bound to anything, and even rebound at runtime for specific instances. There is no difference between access and property and calling a function - it’s all messages. Smalltalk and Objective-C (before dot operators) don’t even have different syntax for data fields and functions calls. Ruby’s no arg messages are similar.

Most of the time that distinction doesn’t matter. But writing things like wrappers and proxies becomes trivial. A object can forward any message it receives, and if it sees on it wants to intercept, it can do that easily. Most of the time modifying existing programs and frameworks can be as easy as rebinding some logic to something that wasn’t part of the original program.

This comes at the cost of some runtime performance, and possibly some complexity. The elegance outweighs those, imho.

Yeah some of my most useful AI tooling are skills created via a “role play session”. Basically brain dumping to the agent and telling it to ask questions and figure out how to accomplish a task, then distilling it into a skill at the end which is much tighter and evidence based from the actual problem solving session

This was very insightful. I've only just begun playing with some agent workflows and building out documentation to help it navigate my code base. Asking it to give me the top 10 unanswered questions from analyzing the docs and code was very useful.

YAGNI is the best tool in your toolbox for AI agents. Dont build out what you think will be useful, layer things into your AI toolbox as they prove they are needed. Especially for claude, running `/init` ends up with a lot of really unnecessary/hallucinated info. Keep it all simple and layer on top.

I think golang is likely a context rot issue. Sure its straightforward but the verbosity ends up affecting long horizon performance

Anecdotally Ive had decent output on a golang project but an incredible amount of duplication that I have to explicitly catch and have it correct.

Oh man I love this so much. The prompts made me laugh so hard. Great experiment.

Holy crap is this underselling how poorly this announcement is structured. Not only does it not provide clarity, it words things in such a way that it just begs more questions. “There are no changes for now”....

Alternatively:

Have you actually used these tools?

They work like magic but are not magical. Theres a skill in using them well, and getting good output. Its not just an automatic free lunch button. Good engineers become great engineers and the gap widens, as juniors/outsourcing gets pushed out of the market.

I do this to remote deploy and it works fine even from my mac

> nix run nixpkgs#nixos-rebuild -- switch --flake .#my-flake-target --target-host nixos@$192.168.x.x --sudo --ask-sudo-password --no-reexec

I also was looking for specific info on the evals, because I wanted to see if they were separately confirming that shoving the skills into the main context didnt degrade the non-skills evals. Thats the other side of skills other than ability to the thing, they dont pollute the main context window with unnecessary information.

I have a couple skills invoked with specific commands ('enter planning mode' and 'enter execution mode') and they have never failed to activate. Maybe make the activation a very rigid phrase and not implied to be a specific phrase.