minimaxir

4 ms·

1.
by minimaxir 5d ago
-2 If OpenAI and Anthropic go bankrupt like everyone romanticizes, it will just lead to a market contraction, not a bubble burst. Those two are not the only AI players.
2.
Writing Rust code faster than SotA by asking agents to make the code faster(minimaxir.com)
1 points by minimaxir 5d ago|0 comments
3.
by minimaxir 5d ago
Good because I don't have to do annoying things like removing stop words or stemming anymore.
4.
by minimaxir 6d ago
Calling an article from the Wall Street Journal AI generated is a stretch.
5.
by minimaxir 7d ago
A new image at a technical level , but all modern autoregressive image models are skilled at only changing what is absolutely necessary if asked to make edits.
6.
by minimaxir 7d ago
> Why would someone want it to be unrecognizable as AI? If we operate under the assumption that the AI generated posters are bad (which this HN thread is), then making them not look like AI generated posters would be directionally towa
7.
by minimaxir 7d ago
If given an underspecified option, yes. Which is why it's good to specify options.
8.
by minimaxir 7d ago
Satire usually has a point or wit (e.g. https://news.ycombinator.com/item?id=49610229 ). This is just a find/replace from "AI" to "SHIT".
9.
by minimaxir 11d ago
LLM caches only work for 100% input matches for all previous text, which will never happen outside of a single session. Things such as system prompts are already cached.
10.
by minimaxir 12d ago
Snow Leopard was one of the most celebrated Mac OS X updates for a reason.
11.
by minimaxir 12d ago
Safari does in fact feel snappier.
12.
by minimaxir 13d ago
Embeddings are very very much used for the use cases mentioned, they're just not sexy to blog about.
13.
World of Warcraft: Forever(worldofwarcraft.blizzard.com)
9 points by minimaxir 14d ago|0 comments
14.
by minimaxir 14d ago
You could put in your AGENTS.md/CLAUDE.md for it to use polars. Modern LLMs use it just fine.
15.
by minimaxir 14d ago
What makes a man turn neutral? Lust for gold? Power? Or were they just born with a heart full of neutrality?
16.
by minimaxir 14d ago
With the exception of data.table (which has a painful API), anything else in R will be much slower. https://duckdblabs.github.io/db-benchmark/
17.
by minimaxir 14d ago
You can do to_pandas() and from_pandas() which doesn't have much overhead, or to_numpy() directly which is typically zero-copy and thus no overhead: https://docs.pola.rs/api/python/stable/reference/d
18.
by minimaxir 14d ago
It's subjective, but IMO polars's APIs are better and than pandas's for ETL. Notably, pandas ported over polars's use of .col() for the 3.0.0 release: https://pandas.pydata.org/docs/whatsnew/v3.
19.
by minimaxir 14d ago
That inertia is not a good thing, and it's partially why there's stagnation in data science. Polars is more than mature enough in both documentation and resources for it to be a daily driver.
20.
by minimaxir 14d ago
At my work I had convinced the ML pipeline engineers to switch from pandas to polars for even small ETL pipelines and there were notable performance gain with better CPU/memory utilization. If a library is performant at large datasets,
21.
by minimaxir 14d ago
It's been a while since I've seen an actual data science post submitted to Hacker News: both because AI has superset a lot of DS tasks (e.g. vector embeddings), but also because not much new has happened in DS. Polars has been aro
22.
by minimaxir 14d ago
> in this economy is the upsell of a new phone every year really justified iPhones are luxury goods ( https://en.wikipedia.org/wiki/Luxury_goods ), not essential goods that need to "justify" a price increase
23.
by minimaxir 14d ago
Creating a throwaway account to "own the left" is substantially more cringe.
24.
by minimaxir 14d ago
The logic is supposed be that "someone else did the same thing as I did, so what I did was correct and your criticism is therefore wrong". It's on me for randomly checking comments to see if people are acting in bad faith. So
25.
by minimaxir 14d ago
People submitting posts to HN automatically has been a thing; the rise of AI agents hasn't changed that and has not affected HN's "immune system". People who need agents to growth hack tend to have crappy sites that alre
26.
by minimaxir 14d ago
If you're going to mock me by repeating my own words on a different HN thread ( https://news.ycombinator.com/item?id=49665394 ), at least make it make sense because this is not analogous other than the Claude-esque desig
27.
by minimaxir 14d ago
Your copy and design is very obviously LLM-generated and full of technical fluff, which unfortunately distracts from any novel technical merits. That's the double-edged sword of marketing, and why simple high-signal copy and examples a
28.
by minimaxir 14d ago
How will GPT-5.6 Sol and GPT-6 Sol be differentiated?
29.
by minimaxir 14d ago
Just because LLMs can write copy for webpages doesn't mean they should . A landing page should not warrant a tl;dr.
30.
by minimaxir 14d ago
If this is accurate, OpenAI's product differentiation will become even more confusing.

More ›