serengetti
serengetti on Hacker News
this is the only way to judge a company.
Posted by
serengetti on Wednesday March 20, 2024 @08:51AM
from the youtube.com dept.
I've been suckered into that tarpit myself. It's a gruelling and mostly pointless process that is designed to suck the soul out of you and crush any hope of ever getting a job there. It starts almost inviting, but gets progressively disparaging and obscure. After the 6th round of interviews I got zero feedback other than "other candidates were better suited for this position".
It's an utter nightmare that any sane person should avoid.
After following all the steps, I see this error:
AttributeError: module 'mpmath' has no attribute 'rational'
i'm confused, in the end, did your documents get processed?
When you've applied in the past, did you get any useful feedback, after your application was rejected?
it's a genuine question, i'm curious if they provide anything actionable or a generic "we found other applicants more in line with our values"
the author only provides vague statements such as "it was a pleasant experience" or "slightly reduced exercise", and ends it with "read about this topic".
so there's no question of reproducibility since there is no method to reproduce.
this is the equivalent of a scientist submitting a paper for publication with one sentence in Materials & Methods section: "we will let the readers infer the exact steps based on our conclusions"
using that same logic I can say "I cured my pancreatic cancer, read the science around this topic to see how"
judging by the article, he'll be selling perpetual motion devices and cancer cure in the next 3-6 months.
entire article: "it's trivial to do, anybody can do it, look we did it for other things. it only cost me 400$"
HOW? "trust me bro"
Posted by
serengetti on Friday February 23, 2024 @01:09PM
from the cbc.ca dept.
that looks awesome. which LLM are you using?
it's the same concept as google maps reviews. a restaurant may not want those comments or have any way of moderating them, but that's what helps to maintain good quality, at least in theory.
probably fell out of the window of his ground-floor prison cell?
try odysee.com, it seems to be a better platform than YT or FB
that's only an app, it would be much easier to use a browser plugin instead, i wonder if they have an API to hook into?
it actually wouldn't cost the manufacturers that much, 90% of vulnerabilities are in software, so skilled developers could address a vast majority of the problems.
car companies don't want to do that, because that means they won't be able to have backdoor access to the car when they want to disable a "pay-per-use" feature, or have some proprietary service done that nobody else is allowed to do.
There even was an entire project Looking Glass in java around it, I thought it was years ahead of its time.
if someone has money, they can buy an asset, mark up the price and resell it.
how do you think walmart, or literally any other store make their money?
that's called capitalism, if you don't like it, move to a socialist country, nobody is forcing you to participate in it.
We wouldn't want to force car makers to hire actual engineers and pay them a living wage to make the cars secure, that would be too much; let's just make crime illegal, that'll solve our problems.
this is what happens when government tries to solve a problem. by the same token they can solve homelessness by making it illegal to sleep on the street.
my uncle got hit by lightning once and he's fine.
want to start a lightning kite-flying club?
Here's an example of an article that doesn't seek validations and where lessons would be obvious:
I developed an ACME service, initially featuring capabilities A, B, and C, and charging $X for it. Over the course of 18 months, through strategic outreach and organic growth, the service successfully attracted 75 corporate clients.
Subsequent to this period, I introduced a more affordable pricing option (for $Y), offering exclusively feature A. This was followed, two months later, by the introduction of feature D, alongside the discontinuation of feature B. Further refining our service offering, a month later, I launched an elevated pricing tier ($Z) that included an additional service, E. However, within four months of this adjustment, the service experienced a cancellation from seven clients. This trend continued, with 28 more clients discontinuing their service six months later, and an additional 40 clients cancelling over the next six months after continuous adjustments to the feature offerings across different pricing tiers. The feedback from departing clients was minimal, with the predominant commentary highlighting that the frequent changes were disrupting their business operations.
here's an unpopular opinion: stop using their "free" services and start self-hosting or paying for 3rd party ones.
the entire article feels like attention-seeking and a bid for external validation.
* what does it mean "experimenting with features and pricing"?
* what was the exact feedback beyond "it disrupted our business"?
* "I gleaned invaluable lessons on how to manage clients" what are those lessons??
what a waste of time!
TL;DR: This entire article is basically regurgitating the lobbyists' party line: "average consumer doesn't want to do it and it's too dangerous for just anybody, we need to keep control!"
How about manufacturers stop suing anyone trying to repair their own equipment? Or how about they stop serializing bloody magnetic proximity sensors?? How is that related to safety and security?
Stop this corporate doublespeak.
Posted by
serengetti on Monday January 29, 2024 @10:06AM
from the github.com dept.
If you use Pinecil v2 soldering iron, you should try this interface for controlling it.