hcks

4 ms·

1.
by hcks 4mo ago
PSA: not making safe environments where you can skip all permissions and instead wasting time monitoring agents == incompetence
2.
by hcks 9mo ago
Guy who made millions selling ads has a meltdown over one (1) spam email
3.
by hcks 2y ago
I know!! Also we could have given an IPhone to 500 million of people for the amount!! It’s such a waste to think they’re investing it in the future instead
4.
by hcks 2y ago
At this point why not go low-tech and use healing crystals?
5.
Polling by asking people about their neighbors: When does this work?(statmodeling.stat.columbia.edu)
2 points by hcks 2y ago|0 comments
6.
by hcks 2y ago
Well either volume weighted or randomised then
7.
by hcks 2y ago
Could you please list say the top 5 Nobel worthy achievements of this backlog
8.
by hcks 2y ago
You're right, Meta shouldn't invest in a breakthrough technology. They should focus on what really matters: delivering short term value to shareholders
9.
by hcks 2y ago
Round of applause for this useless unsubstantiated comment
10.
by hcks 2y ago
This outrage talk is ridiculous, you don’t have the morale high ground to judge people, they didn’t break anything, taking VC money isn’t bad, the incident was caused by sore losers looking for drama, etc etc
11.
by hcks 2y ago
This is actually an incredibly stupid thing to say which reflects poorly on NASA management
12.
by hcks 2y ago
Hackernews when we may be on the path of actual AI "meh I hate this, you know what’s actually really interesting? Manually writing tests for software"
13.
by hcks 2y ago
And by nowadays you mean since ChatGPT got released, that is less than 2 years ago (e.g. a consumer preview of a frontier research project). Interesting.
14.
by hcks 2y ago
Okay but how big of a sample size do we even actually need for word frequencies? Like what’s the goal here? It looks like the initial project isn’t even stratified per year/decade
15.
by hcks 2y ago
> I recently built a Brainfuck compiler and TUI debugger Highly representative of what devs make all day indeed
16.
by hcks 2y ago
> HN on LLMs "meh not real reasoning, in fact it's exactly the same thing as ELIZA" > HN when someone makes a trivial LLM-wrapper (but cynic!! I hate corporate ahah): 9999 comments 9999 points "this is the best thi
17.
by hcks 2y ago
> The young ones don't understand why we oldies roll our eyes at hearing the 13th iteration of "this is ground breaking new idea" Why not constructively contribute by explaining what could be different for the idea to work
18.
by hcks 2y ago
Yes software dev has 10 good years left
19.
by hcks 2y ago
“We thought academia was not soul crushing enough so from now on you will additionally spend 10 hours a week replicating dumb papers from 1993”
20.
by hcks 2y ago
The introductory example is quite illuminating. No theory behind, just a random hypothesis tested (by comparing the preferences of 10 fishes from populations separated from as little as 50 meters, what effect size did the authors expect?),
21.
by hcks 2y ago
Can someone mourning can me one single valuable link that will be lost?
22.
by hcks 2y ago
What if psychedelics are actually not gates to an enhanced consciousness for the happy few? What if actually your teacher was right, and they’re just dumb intoxicants that have no upside and a small chance of ruining your life?
23.
by hcks 2y ago
LangChain is a critical thinking test and orgs using it are ngmi
24.
by hcks 2y ago
Don’t waste your time, it’s been around since GPT3, and had no results so far. Also notice how no frontier lab is working on it.
25.
by hcks 2y ago
Where all the “that’s an America problem, here in the EU…” posters?
26.
by hcks 2y ago
Civil servants are indeed well known for their non-conformity and appetite for freedom
27.
by hcks 2y ago
It’s not been audited by CIA approved programmers
28.
by hcks 2y ago
"I refuse categorically to entertain fun maths problems."
29.
by hcks 2y ago
Code interviews are great: - filter out low-IQs - filter out people difficult to work with that refuses to take them - filter out people that have no interest whatsoever in computer science
30.
by hcks 2y ago
In the 1D case the single point will be the best estimator for the mean no matter what the variance is

More ›