avereveard

4 ms·

31.
by avereveard 3mo ago
Seems hella inefficient. Better method start to realizing that everything that every program do is data transformations and or movement Then you ask llm to subdivide data in a tree along the domain model, classifing streaming vs storing nod
32.
by avereveard 4mo ago
I really dislike opus 4.8 it rarely compete things and prefer to waste tokens making lists of things that are missing. When stuck or need input it words the challenge at length without conveying anything useful for decision making, and quit
33.
by avereveard 4mo ago
I hate loud ads as well as anyone else and I welcome this resolution but I would not treat the challenges regulation poses as simplisticly as this. There is a lot of research in increasing loudness without increasing decibels, especially fo
34.
by avereveard 4mo ago
> Anthropic is subsidizing their enterprise customers by up to 40 times, and OpenAI up to 70 times might as well be the other way around with non subscribed token being 50x overpriced, or any combination thereof also uber was non profita
35.
by avereveard 4mo ago
https://en.wikipedia.org/wiki/Prohibitory_traffic_sign#/medi... yeah?
36.
by avereveard 4mo ago
There literally are "no u turn" signage where you are supposed not to do that. They literally put up signs for it. It is not glowing in the sky, and it doesnt need to be, and doesnt help making a point strawmanning it.
37.
by avereveard 4mo ago
Until they reverse on a highway and kill a family. Being able to drive isn't where parent poster put the bar
38.
by avereveard 4mo ago
yeah been away from a while. thanks!
39.
by avereveard 4mo ago
Does BAR have chicken maps?
40.
by avereveard 4mo ago
most merge improve a small subset of "feeling" benchmark (too small, too specific, or out of distribution) and tend to show degradation on actual benchmark, with especially punishing result on long chain benchmarks. also only work
41.
by avereveard 4mo ago
I use glm for all code investigations and top level system design of all kinds, and then present finding to confirm and act upon to opus. everything that burns token goes there. the finding aren't always accurate, but it saves ton of o
42.
by avereveard 4mo ago
*Anthropic requires it
43.
by avereveard 4mo ago
Metadata does not require data and conflating them doesn't help your point
44.
by avereveard 4mo ago
The right level of recording of non crime is zero wdym reduced?
45.
by avereveard 5mo ago
Same, even opus favor short term solution and scripts with a billion flags that constabtly require rescanning to understand how to launch it is a constant struggle to get it to build sane default and reusable scripts that run with minimal p
46.
by avereveard 5mo ago
But they are, especially those with batteries
47.
by avereveard 5mo ago
Fiat Multipla level design blunder
48.
by avereveard 5mo ago
Also nobody talks enough about the fact that workforce is effectively cut out from the means of productions. Even with the capital at hand blackwell cabinets are all sold out, contracted to the big providers. There are paralles to the indus
49.
by avereveard 5mo ago
The demand doesn't necessarily double.
50.
by avereveard 5mo ago
System security is not a human value. Access key rotation effective immediately is a compliance requirement, and completely orthogonal to human decency, which is delivered trough garden leave or severance, not extended system access
51.
by avereveard 5mo ago
https://arxiv.org/pdf/2508.09101 tldr 2% average point lost on Rust compared to python, gap vary by model, go has a better upper bound but opus had it 3% below python. benchmark is a bit old but research on why is ther
52.
by avereveard 5mo ago
All the control plane . Data plane is distributed and roles using iam to access resources can still do so during a control plane outage.
53.
by avereveard 5mo ago
You don't wait. You run multiple independent incremental feature in parallel, while also running a code review, which will create the next set of tasks while you or the llm think up the feature to add after.
54.
by avereveard 5mo ago
It's surprisingly hard to do the compiler or cpu may see a write without a read and optimize it away. Windows has a SecureZeroMemory and a few other barrier primitives but not all languages reach to it
55.
by avereveard 6mo ago
For better and worse tho.
56.
by avereveard 6mo ago
eh, good programmer are goal oriented, today SOTA models still need for the most part step by step guidance, so there's a gap still. the AGENTS.md pieces that pin specific tool-call shapes or force chain-of-thought before action are co
57.
by avereveard 6mo ago
"astounding how much the harness matters" is the right read and it should be the lasting one. the model is rentable, the prompts are rentable, the benchmark numbers are mostly a function of the harness around them. swapping Gemini
58.
by avereveard 6mo ago
"Minimal infrastructure, max LLM freedom" works great for personal automation. The same shape under enterprise security review collapses on the question of what cannot happen, which is exactly what the prompts-and-vibes school d
59.
by avereveard 6mo ago
as knowledge is commoditized the bar for junior raises, what was advanced math research a century ago is now undergrads' homework. I don't see why code is so special in that regard that cannot progress beyond artisanship.
60.
by avereveard 6mo ago
the ecosystem obsession with public benchmarks comes from the fact that running benchmark costs, and labs don't test on any given private benchmark but yeah you're correct anyone optimizing for public-bench rank instead of their o

More ›