
Programmer. @CodeWebChat Life's most persistent and urgent question is, “What are you doing for others?”
Poland
Joined November 2012
- Tweets2.8K
- Following904
- Followers244
- Likes4.3K
Pinned Tweet
A sessionless AI coding
- agent selects task-relevant files on a file tree
- you select examples for guidance
- editing, asking questions without tool calling
- great prompt cacheability by smart context ordering
- no context accumulation & bloat
- clear intent = less 'thinking'
2
210
Robert Piosik retweeted
If our small minds, for some convenience, divide this universe, into parts — physics, biology, geology, astronomy, psychology, and so on — remember that nature does not know it!
It could be that user requirements for model capabilities are artificially inflated by poorly engineered workflow.
Too natural to stay in a convo... Too natural to bloat context window with iterations...
It's like we unwittingly bury intent and dig it out in every turn 🥲
2
116
Robert Piosik retweeted
everybody’s going to build their own software just like everyone builds their own houses and cars
Robert Piosik retweeted
Replying to @AgustinLebron3
Not really observations that others before me haven’t made, but:
- Despite the seemingly magical nature of LLMs, reflection over a >3 month timescale suggests my total productivity hasn’t increased by over 100%, or perhaps even by over 50%, and a lot of time is actually wasted because LLMs enable me to spend time on gratifying but low-productivity tasks that in the future turn out to not be useful
- Also, capabilities are incredibly spiky and highly correlated with the degree of investment poured into them, which my earlier tweet about math benchmarks implicitly points out
- From the above, it seems that the nature of LLM intelligence is wildly dissimilar to that of human intelligence and we won’t trivially get to something superior to human intelligence in all important respects just by scaling up existing approaches with various tweaks; even if AGI Is eventually achievable, this implies a significantly longer timeline
- Benchmark progress is almost definitionally guaranteed to happen because the process of constructing a benchmark is a direct precursor to the process of constructing a training dataset used for hill climbing that benchmark, but the scope of what can be captured in a benchmark is (at least for now) grossly lacking in terms of its relevance to real-world work, with maybe several limited exceptions
- Progress seems highly gated by data but the nature of model training means that each “next dataset” is significantly harder to assemble than what preceded it; some wins are possible through synthetic methods but those feel more like “patching up gaps” than “pushing the frontier forward”
At a higher level, I guess I’d say there’s a sort of refusal to think carefully about what models are or are not useful for in a rigorous way which I find personally quite annoying, and instead a reliance on some nebulous notion of being “AGI pilled” as a replacement for serious thought. I think people are very quick to anthropomorphize LLM intelligence because humans communicate through words and we infer the intelligence of human counterparties through comprehension of their language, but this leads them to wrong conclusions; for example if we observe that a new model proved some incredible mathematical theorem, some will say, “well, don’t we have AGI now, huh?” But to me, it’s actually more like, “well, given how hard it would have been for a human to do these mathematics, and given the limited economic effect of LLMs upon the world so far, isn’t it actually a negative datapoint vis-a-vis the generality of LLM intelligence?”
Robert Piosik retweeted
Does one successful run prove that you wrote a good prompt? No. It proves only that one possible execution happened to satisfy your intent.
Selecting context files makes AI coding 'too cheap to meter'.
References, imports and parent folder selection 'is all you need'. 🤓
1
111
Robert Piosik retweeted
Replying to @addyosmani
Friction also forces simplicity.
"Simplicity is a prerequisite to reliability" ~Edsger Dijkstra
with agents, it's very easy to create unnecessary complexity and I bet serious companies will start to worry
enforce friction - expect simplicity - get a hit on reliability 😉
I'd say these things are true if one can't select context files required to implement the prompt 🤓
If you’re not reading the code, whether explicitly or through agentic inquiry, one or more of these is true:
○ You’re a beginner
○ Software is throwaway
○ You’re prototyping
○ You have no users / revenue
○ You’re taking on debt & risk
○ Your problems are basic
And btw. All of this is fine. But the reality is that models are still not at the “full autonomy” stage yet.
They make rookie mistakes, they go down bad architectural paths. I just had the best model in the world add a nonsensical 700ms delay to “settle” something and it told me “you’re right, I was cargo-culting” 🤨
I am on the camp that this need will diminish more and more. Most code is indeed going to be assembly-like. But we also have the global internet and software infrastructure riding on these models and narrative, and we have to respect that.
128
Why are models still bad at writing? Because they can't make intelligent word choices.
It's the same with code, unless provided with examples.
They don't work with text, because creativity is expected, but with code, examples are like fixed guidelines 🤓
3
80
It's interesting to see push for cloud agents from every corner.
Technically it's impossible to train a model that understands interconnected logic and does architectural house-kepeing.
This great limitation opens to incredible fragility whenever business requirements change.
Everything has changed for us with orbs.
My biggest struggle right now is figuring out how to make you see what we see.
So I sat down and wrote about it:
ampcode.com/notes/what-i-wan…
123
One of agentic penalties are 'unmanaged' identifiers.
Later, when asking a model for explanations, it could use vocabulary that is distant to your ways of thinking.
1
2
82
Quality software is made of small, carefully planned steps. The more fine-grained you go, the more consistent with its outputs the model becomes.
With consistency, there is trust.
Trust allows flow state to happen.
Flow state is happiness.
1
3
128












