cloud kid. devops dude. all for the glory of God † not important nor clever enough for anyone to claim my thoughts as their own.
127.0.0.1
Joined March 2019
- Tweets2.1K
- Following511
- Followers56
- Likes2.8K
jack://nemitz retweeted
The same people who used to just copy and paste a solution from stack overflow without caring to understand it are vibe coding now.
The same people who genuinely care about understanding and owning the problem and code are still doing that today with AI
jack://nemitz retweeted
Managing a lot of agent is really hard. If people say it's easy it's because they actually suck at what they do.
jack://nemitz retweeted
a weird inversion with LLMs is the models improve faster than the tinkerers
when i see people with custom workflows and setups they're all addressing problems that don't exist anymore
the person naively using vanilla codex is more likely to be experiencing state of the art
jack://nemitz retweeted
In a phrase, "make bad code unrepresentable"
My hot take from chatting to @poteto is that we should use MORE abstractions in the AI age
You can use them (combined with harsh lint rules) to reduce the design space available to the agent and constrain them only to good decisions.
Combined with the fact that high-leverage abstractions let you do more with less code - so, more token efficient.
Plus, unwinding the damage from a bad abstraction is much cheaper with agents.
This runs counter to a lot of folks thinking that agents just want to read the raw code. They can, but they're not maximally efficient that way.
Be braver! Design abstractions.
jack://nemitz retweeted
The "whiteboard defense:" I should be able to pull you aside at any moment and ask you to explain any customer-facing system you've shipped. You should be able to clearly explain how it works and defend the decisions you made. This is my benchmark for responsible AI usage.
I don't expect line-level familiarity with the code. I don't care if you remember the exact function name or implementation detail. You may not even know it. I don't care.
But if I ask "why did you do X instead of Y?", "what happens if this actor behaves maliciously?", "what data structure did you use here and why?", or "where does this fail?" you should be able to answer confidently.
For PoCs, demos, experiments, whatever: I don't care. Generate 100% of it and understand none of it. Speed over quality every time in those specific scenarios.
But if you're shipping customer-facing work, you can't be shipping things you don't understand at a high level.
jack://nemitz retweeted
the general public has been getting constant waves of fear and confusion from ai companies
i thought the whole "you will be replaced" thing was as bad as it gets but now it's "you will literally die"
the defense is always "it would be irresponsible of me not to warn people of the risks"
but all that's happening here is the same thing that always happens: a group of people had the thought that their cause is righteous and moral
and they're incrementally justifying worse and worse behavior and entirely incapable of reflecting on how they might be the problem
jack://nemitz retweeted
the fact of the matter is that a sub-frontier open weights model like GLM-5.2 is more than capable of transforming a professional software development workflow.
the unattended workflows will eventually fail. we just need to wait for the hype to die down and the responsible professionals to feel the pain of losing cognitive ownership
then we’ll see a wave of “why we abandoned autonomy and focused on cognitive ownership” posts. most likely describing processes similar to lode coding and its cousins.
it’s already over for the frontier labs when it comes to software development. today’s open weights are enough.
jack://nemitz retweeted
My "plan mode" is now:
<Rambling 2 minute dictation> + "Repeat back what I just said in basic, structured language and explain how you're going to accomplish this to completion."
jack://nemitz retweeted
If you give me a drawer, I will eventually fill it with cables
If you give an agent a markdown file, it will eventually fill it with:
- unnecessary implementation details
- session-specific observations
- stale docs
jack://nemitz retweeted
Man, I love this stuff.
If I want to code by hand, I code by hand. If I want to code with AI, I code with AI. If I want Omarchy, I use Omarchy. If I want to build my own rice from scratch, I do that too. Rust one day, Java the next.
Don't forget why you got into tech in the first place. Remember that joy you felt writing hello world for the first time, or discovering that distro that just clicked, or trying out a keyboard that felt perfect?
It was euphoric. So why try to take that away from someone else?
Have opinions, argue about what's better, whatever. But I'll never understand being so judgmental toward someone for enjoying tech differently.
Some of y'all just need to roll one up and watch Surf's up or something.
