Sabitlenmiş Tweet
The strongest teams, orgs, and companies are ones that value the engineer the most.
True or false?
English
Armen Shimoon
3K posts
Armen Shimoon
@ArmenShimoon
PE @ Amazon Alexa | applying agentic engineering to all aspects of my job
USA Katılım Haziran 2011
3K Takip Edilen2K Takipçiler
Love the division. I call this blue jobs vs pink jobs. Typically blue jobs are outside and pink ones are inside, but not always. Anything nasty or heavy is something I just do for my wife, regardless of location. Killing bugs, wiping up dog puke, moving furniture, and taking garbage from the kitchen (all the way) are all examples.
Also if she needs some extra help I just automatically help with stuff I don't typically do. Because we are in sync, on the same team, and both work hard.
In the end we all know whether a job is blue or pink and we just do them. Not keeping score, but appreciating and helping the best ways we can.
English
Taking out the trash, because it is partly an outdoor activity, generally falls under the man’s purview. I now have a son old enough to delegate that chore to, but the arrangement is that my wife ties the bags and places them by the door, and I take them out. In this case, my son now takes them out.
Sometimes the bags are quite heavy, so this is a practical kindness I can do for her. In general, outdoor tasks fall to the man and indoor tasks to the woman, and this is a shared task that sits between the two.
English
The best thing I ever did for my marriage was to tell my wife that I would no longer be doing any chores in the house. No cooking, no cleaning, nothing.
I was going to focus on making money so the family had plenty of resources, and she would focus on the domestic responsibilities. Everything has been better ever since.
A clear division of responsibility reduces marital conflict.
English
With agentic engineering, it's not about how much you use coding agents.
The only thing that matters is whether the tokens you produce are more valuable than the cost of your input tokens.
You wanna spend $100 on input tokens to produce a tool that makes $200. That's it.
English
@karpathy Rust is definitely nice, but what I've learned is that very quick compile times are far more important than raw runtime performance. I've gravitated toward Golang for this reason, still high performance, builds and tests super quick. Fast iteration loops are invaluable.
English
I think it must be a very interesting time to be in programming languages and formal methods because LLMs change the whole constraints landscape of software completely. Hints of this can already be seen, e.g. in the rising momentum behind porting C to Rust or the growing interest in upgrading legacy code bases in COBOL or etc. In particular, LLMs are *especially* good at translation compared to de-novo generation because 1) the original code base acts as a kind of highly detailed prompt, and 2) as a reference to write concrete tests with respect to. That said, even Rust is nowhere near optimal for LLMs as a target language. What kind of language is optimal? What concessions (if any) are still carved out for humans? Incredibly interesting new questions and opportunities. It feels likely that we'll end up re-writing large fractions of all software ever written many times over.
Thomas Wolf@Thom_Wolf
Shifting structures in a software world dominated by AI. Some first-order reflections (TL;DR at the end): Reducing software supply chains, the return of software monoliths – When rewriting code and understanding large foreign codebases becomes cheap, the incentive to rely on deep dependency trees collapses. Writing from scratch ¹ or extracting the relevant parts from another library is far easier when you can simply ask a code agent to handle it, rather than spending countless nights diving into an unfamiliar codebase. The reasons to reduce dependencies are compelling: a smaller attack surface for supply chain threats, smaller packaged software, improved performance, and faster boot times. By leveraging the tireless stamina of LLMs, the dream of coding an entire app from bare-metal considerations all the way up is becoming realistic. End of the Lindy effect – The Lindy effect holds that things which have been around for a long time are there for good reason and will likely continue to persist. It's related to Chesterton's fence: before removing something, you should first understand why it exists, which means removal always carries a cost. But in a world where software can be developed from first principles and understood by a tireless agent, this logic weakens. Older codebases can be explored at will; long-standing software can be replaced with far less friction. A codebase can be fully rewritten in a new language. ² Legacy software can be carefully studied and updated in situations where humans would have given up long ago. The catch: unknown unknowns remain unknown. The true extent of AI's impact will hinge on whether complete coverage of testing, edge cases, and formal verification is achievable. In an AI-dominated world, formal verification isn't optional—it's essential. The case for strongly typed languages – Historically, programming language adoption has been driven largely by human psychology and social dynamics. A language's success depended on a mix of factors: individual considerations like being easy to learn and simple to write correctly; community effects like how active and welcoming a community was, which in turn shaped how fast its ecosystem would grow; and fundamental properties like provable correctness, formal verification, and striking the right balance between dynamic and static checks—between the freedom to write anything and the discipline of guarding against edge cases and attacks. As the human factor diminishes, these dynamics will shift. Less dependence on human psychology will favor strongly typed, formally verifiable and/or high performance languages.³ These are often harder for humans to learn, but they're far better suited to LLMs, which thrive on formal verification and reinforcement learning environments. Expect this to reshape which languages dominate. Economic restructuring of open source – For decades, open-source communities have been built around humans finding connection through writing, learning, and using code together. In a world where most code is written—and perhaps more importantly, read—by machines, these incentives will start to break down.⁴ Communities of AIs building libraries and codebases together will likely emerge as a replacement, but such communities will lack the fundamentally human motivations that have driven open source until now. If the future of open-source development becomes largely devoid of humans, alignment of AI models won't just matter—it will be decisive. The future of new languages – Will AI agents face the same tradeoffs we do when developing or adopting new programming languages? Expressiveness vs. simplicity, safety vs. control, performance vs. abstraction, compile time vs. runtime, explicitness vs. conciseness. It's unclear that they will. In the long term, the reasons to create a new programming language will likely diverge significantly from the human-driven motivations of the past. There may well be an optimal programming language for LLMs—and there's no reason to assume it will resemble the ones humans have converged on. TL; DR: - Monoliths return – cheap rewriting kills dependency trees; smaller attack surface, better performance, bare-metal becomes realistic - Lindy effect weakens – legacy code loses its moat, but unknown unknowns persist; formal verification becomes essential - Strongly typed languages rise – human psychology mattered for adoption; now formal verification and RL environments favor types over ergonomics - Open source restructures – human connection drove the community; AI-written/read code breaks those incentives; alignment becomes decisive - New languages diverge – AI may not share our tradeoffs; optimal LLM programming languages may look nothing like what humans converged on ¹ x.com/mntruell/statu… ² x.com/anthropicai/st… ³ wesmckinney.com/blog/agent-erg… ⁴ #issuecomment-3717222957" target="_blank" rel="nofollow noopener">github.com/tailwindlabs/t…
English
@simeonGriggs I built a skill manager skill that does exactly this github.com/ashimoon/claud…
English
Can anyone explain why we're not installing skills from npm so we actually get versioning and updates?
Downloading plain text files to your project exactly once feels insane.
English
@_luki222 @GenAI_is_real Thats where the human driver comes in
English
@ArmenShimoon @GenAI_is_real There are no gains from AI in planning. Its too complex, even impossible to fit relevant info in context. Sure you can tell it to design some system that was previously already scoped out by a sr eng, but AI can't design or plan end to end coordinating multiple teams and products
English
FAANG is literally panicking refactoring because human code is now the bottleneck. But honestly, monorepos won't save them from the infinite spaghetti code agents are about to dump. OAI already has internal tools for this that make Bazel look like a toy. The era of human "senior engineers" is ending faster than you think @karpathy @sama
Samswara@samswoora
Rumor is FAANG style co’s are refactoring their monorepos to scale in preparation for infinite agent code
English
@_luki222 @GenAI_is_real Companies need to refactor their organization and processes, not code. Agents can refactor the code just fine, but overbloated and slow moving people will continue to be in the way, preventing AI unlocks and gains from being realized.
English
@GenAI_is_real FAANG eng here. There is 0 reason to believe this. There is no refactoring happening. Typing code isnt the bottleneck and never was. The actual bottlenecks are crossteam alignment, system design and organizational complexity. Humans produce as much spaghetti as AI agents anyway.
English
@ThePrimeagen In your analogy is the ore being dug by shovels or is the ore being used to make shovels?
English
reason being is the following
1. right now its most profitable to be in the process of making ore for the shovels
2. when that flips and the process of selling shovels is more profitable, nvidia will own all the ore and deep seek taught us that brain draining is real
3. they would be able to scale up as fast as energy production and their own production allows them
at the end of the day i cannot see why companies that create models win, they are a stopping point for those that own the real resource, production of the chips.
anywho, just coding and thinking
ThePrimeagen@ThePrimeagen
new prediction Nvidia will buy OpenAI
English
I use Claude Code for anything beyond trivial terminal commands nowadays.
Its the first thing I install when I spin up new VMs/containers. The rest of the setup, and effectively the terminal work after that -- I run through Claude.
Arpit Bhayani@arpit_bhayani
I do not remember shell commands :) and I needed to kill a process running on port 21079. Instead of struggling to call and running a couple of commands, I built a small utility called `x`. > x kill process running on port 21079 This utility is a pure bash script with no external dependencies (just curl or wget). You describe what you want in plain English after "x", and it generates the shell command and executes it only after getting your confirmation. > x get all the git branches > x delete all pycache and useless python files > x grep all errors from ~/access.log file It supports OpenAI, Gemini, and Anthropic as providers. You just set one API key, and you are good to go. I have open-sourced it, and the repo is linked below. Give it a star. Also, feel free to fork it and tweak it. Installation takes a single command and barely a minute. Give it a spin.
English
With Claude Code I still do rubber duck debugging
Except now I am the rubber duck
English
@alexalbert__ Claude Code:
- if using with tmux and your change your layout, the UI gets messed up
- sometimes does this weird scroll thing when you are selecting from different options. Like 60 seconds of scrolling as you select an option
- some paths show as ../../../../../../../foo.json
English
Reply with all your Opus 4.5 gripes so we can fix everything before our next model
The more specific (including prompts), the more likely we'll be able to fix it!
English
@amyoder The real move here is to bet $250 that he does give the $2k. If he gives, you are up $1750 total (you lost just your wager). If not, you are up $2000.
English
This is the literal definition of Free Money
And here’s exactly how you finesse it
Bet $2,000 on “Yes” on $2k Trump dividend
If it hits, you walk out with $16,000 profit
If it doesn’t hit, you lose your $2k but then Trump gives you $2k anyways
Either way u win
Risk free
NewsWire@NewsWire_US
JUST IN: 90% chance Trump doesn’t follow through on his $2,000 tariff dividend promises — Polymarket
English
@BobLoukas If there are years left in this rally, then why get cute now trying to micro-time it? Seen so many people get left behind in Bitcoin trying to do the same thing.
English
Sold a decent clip of Gold/Silver positions on Friday. Left the miners alone. First selling since taking positions in 2023.
Sold without loss of a trend too, which I try not to do.
But it’s just hard to ignore how extended its become. How much attention it suddenly has. And frankly, some rebalancing was needed after that move. Still well positioned for upside and dry powered for any meaningful correction, as we have years left in this bull.
English
@HansThorLief @aestheticprimal Supplement with butyrate, PHGG, glutamine, resistant starch, zinc carnosine, and two apples per day (pectin).
English
@aestheticprimal Right. Makes sense.
I’ve seen some of there micros being thrown around on here for many years. But never tried.
Any thoughts on the art of getting a specific strain to settle down and get comfortable in there?
English
The gut microbiome strongly dictates social hierarchy in mammals
Simply populating the gut with certain probiotics could literally induce 'dominance' in rodents and other mammals
Examples of these probiotics are:
- Clostridium Butyricum
- L. Reuteri
- L. Rhamnosus
- Akkermansia
English
Looks like solving high estrogen is just too damn easy.
> Kestose
> White button mushroom
> Olive leaf extract
> Androsterone enanthate
On their own or together they all work great.
English
Kestose = the most slept-on gut prebiotic.
People waste time with weak fibers like PHGG or inulin. What a joke.
Kestose has been shown to:
> Explode anti-inflammatory F. prausnitzii
> Boost a unique pro-longevity bile acid
> Increase 5AR → more DHT + better cortisol clearance
> Improve your T:E2 ratio
> Be tolerated even by the “sensitive gut” crowd
> Fix food intolerances (even lactose)
It’s the king of prebiotics. I put it in my TestoShakes daily.
English
@LibertyLog1776 @StefanMolyneux I am defining that nothingness (or everythingness) from which everything came as God, and I do not for a minute pretend to be able to explain its origins or nature.
English
@ArmenShimoon @StefanMolyneux Special pleading. You're saying "everything needs a cause, except for things that don't, and I know what things don't"
English
What is the biggest thing people in the future will look back and be unable to believe that we believed?
English
@LibertyLog1776 @StefanMolyneux God always was.
But yes it begs the question: why something and not nothing? Why is, always was, and always will be God?
English
@ArmenShimoon @StefanMolyneux So....it's God that popped out of nowhere from nothing? That's much more logical.
English
So let me get this straight:
We just broke a 4-year resistance at $4,000, had a healthy move up, and now after the first 10% dip you’re scared and selling everything?
As long as $ETH holds above $3,800, stay long.
This isn’t the end, it’s the beginning.
English