The new forums will be named Coin Return (based on the most recent vote)! You can check on the status and timeline of the transition to the new forums here.
The Guiding Principles and New Rules document is now in effect.
Seizing the means of Structural [chat]eneering
BeNarwhalThe Work Left UnfinishedRegistered Userregular
I thought it was 0-10 for main, 10-15 for backup, and 15+ for rogue
life's a game that you're bound to lose / like using a hammer to pound in screws
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
0
ChanusHarbinger of the Spicy Rooster ApocalypseThe Flames of a Thousand Collapsed StarsRegistered User, Moderatormod
benards is fake news
i have made no fewer than six timely, non-nonsense [chat]s
i am notorious for my austere and minimalist OPs produced within mere seconds of the previous closing
Allegedly a voice of reason.
0
jungleroomxIt's never too many graves, it's always not enough shovelsRegistered Userregular
Having interviewed STEM job applicants I can say that while there is no shortage of STEM graduates, there is a shortage of good STEM graduates.
My company isn't even trying to underpay or offer shit benefits and they've already accepted that they won't be getting google and apple level talent. And yet still, candidate after candidate cannot answer simple questions.
The ones that can answer are typically swimming in offers and take whatever tickles their fancy.
Can you give examples? This kind of thing makes me curious
canny edge detection and building boxes, etc
The canny edge detection story is actually a good example. My company does computer vision, so every applicant for certain positions has taken a computer vision course and listed computer vision as a specialization for a post-graduate degree. So keep that in mind.
Canny edge detection is one of the basic, most commonly taught computer vision algorithms. Its an image transform that takes in a fully detailed photographic image and spits out another image that looks more like a line drawing. Its great for, as its name suggests, detecting edges. It lets a computer greatly cut down the information content of an image, and make the content left over invariant to lighting, shading, etc. It's typically taught in Image Processing or Computer Vision 101 type classes.
I asked an open-ended question of a candidate. "How would you find a cardboard box in an image of a cardboard box?" The question tests a variety of things, but it's very open to discussion. One candidate in their answer mentioned the canny edge detector, but the way they used it in their design seemed a little odd and skipped a lot of steps, so I asked them to go back and explain. They looked at me like I had just ripped their pants down, so I knew I was in for a treat.
Over the course of their explanation, it became clear that they didn't know what the canny edge detector was or what it did or what its inputs and outputs were. These are not hard questions. It takes in a photo, it spits out a line drawing. But they couldn't say it because they didn't know it. They just threw out the words and hoped I'd let them gloss over it. And I would have, if their usage had made any sense.
From there the interview went rapidly downhill. It was clear they didn't have any computer vision experience and they failed the basic C/C++ programming test too. And this is not uncommon. This is representative of about half of the people who make it through the screening process. These are people with nice GPAs and degrees from good institutions. It's really demoralizing.
I mean, if I'm ready canny edge right on wikipedia, it's a grayscale/bluring (reduce the number of tiny sharp edges?) image where they scan over the ints in the bitmap and check for something breaking a threshold between adjacent values, seems pretty straight forward.
If you were doing computer vision as a degree this seems like something you'd be doing every god damned day for research and projects to the point where it'd be second nature.
Is the C++ just like.. testing if you know for/if/else/variables or are you doing OOP, structs, bit packing, etc ?
Having interviewed STEM job applicants I can say that while there is no shortage of STEM graduates, there is a shortage of good STEM graduates.
My company isn't even trying to underpay or offer shit benefits and they've already accepted that they won't be getting google and apple level talent. And yet still, candidate after candidate cannot answer simple questions.
The ones that can answer are typically swimming in offers and take whatever tickles their fancy.
Can you give examples? This kind of thing makes me curious
canny edge detection and building boxes, etc
The canny edge detection story is actually a good example. My company does computer vision, so every applicant for certain positions has taken a computer vision course and listed computer vision as a specialization for a post-graduate degree. So keep that in mind.
Canny edge detection is one of the basic, most commonly taught computer vision algorithms. Its an image transform that takes in a fully detailed photographic image and spits out another image that looks more like a line drawing. Its great for, as its name suggests, detecting edges. It lets a computer greatly cut down the information content of an image, and make the content left over invariant to lighting, shading, etc. It's typically taught in Image Processing or Computer Vision 101 type classes.
I asked an open-ended question of a candidate. "How would you find a cardboard box in an image of a cardboard box?" The question tests a variety of things, but it's very open to discussion. One candidate in their answer mentioned the canny edge detector, but the way they used it in their design seemed a little odd and skipped a lot of steps, so I asked them to go back and explain. They looked at me like I had just ripped their pants down, so I knew I was in for a treat.
Over the course of their explanation, it became clear that they didn't know what the canny edge detector was or what it did or what its inputs and outputs were. These are not hard questions. It takes in a photo, it spits out a line drawing. But they couldn't say it because they didn't know it. They just threw out the words and hoped I'd let them gloss over it. And I would have, if their usage had made any sense.
From there the interview went rapidly downhill. It was clear they didn't have any computer vision experience and they failed the basic C/C++ programming test too. And this is not uncommon. This is representative of about half of the people who make it through the screening process. These are people with nice GPAs and degrees from good institutions. It's really demoralizing.
This is one of those things that I constantly kind of wonder/worry about. Like, if I was ever in an interview scenario like this one for something relevant to my field (let's say something molecular biology based, or hell, even computationally, because I'm sort of a computer scientist, albeit a shitty one).
I'm like, reasonably certain I would flub interview questions like this, but at the same time I know that I can figure a lot of stuff out- my entire postdoc has been me reading a technique on the internet i've never done before, and trying it out to see if it works
but if like, during an interview you asked me, say, how i would use southern blotting to investigate my phenotype of interest I would probably freak out and not be able to answer, despite the fact that A. I could look it up and B. follow the protocol and C. design a well-structured experiment around it
I know they aren't really equivalent scenarios, I'm just wondering how much I actually know, and how much of it was me coasting through like a lot of these candidates
I am forever filled with impostor syndrome
my final words, as I lay on my deathbed, having been a professor for years and years with a bunch of papers to my name will be
vanilla is expensive. Such that some producers will individually brand every bean to help track their shit to prevent theft
big image for detail:
life's a game that you're bound to lose / like using a hammer to pound in screws
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
Posts
If I'd have known I could have just made a rogue chat I'd have made bushcraft chat.
https://pubs.usgs.gov/fs/2003/fs014-03/pipeline.html
who do i choose
I thought it was 0-10 for main, 10-15 for backup, and 15+ for rogue
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
i have made no fewer than six timely, non-nonsense [chat]s
i am notorious for my austere and minimalist OPs produced within mere seconds of the previous closing
It was.
You lack the fortitude and the certainty to act with such boldness, AH!
I create this [chat] by divine right!
I mean, if I'm ready canny edge right on wikipedia, it's a grayscale/bluring (reduce the number of tiny sharp edges?) image where they scan over the ints in the bitmap and check for something breaking a threshold between adjacent values, seems pretty straight forward.
If you were doing computer vision as a degree this seems like something you'd be doing every god damned day for research and projects to the point where it'd be second nature.
Is the C++ just like.. testing if you know for/if/else/variables or are you doing OOP, structs, bit packing, etc ?
@Donkey Kong
not this bridge nerd shit
I choose BOTH
REUNIFICATION IT WILL BE!
I am altering the deal.
Pray that I do not alter it further.
This is one of those things that I constantly kind of wonder/worry about. Like, if I was ever in an interview scenario like this one for something relevant to my field (let's say something molecular biology based, or hell, even computationally, because I'm sort of a computer scientist, albeit a shitty one).
I'm like, reasonably certain I would flub interview questions like this, but at the same time I know that I can figure a lot of stuff out- my entire postdoc has been me reading a technique on the internet i've never done before, and trying it out to see if it works
but if like, during an interview you asked me, say, how i would use southern blotting to investigate my phenotype of interest I would probably freak out and not be able to answer, despite the fact that A. I could look it up and B. follow the protocol and C. design a well-structured experiment around it
I know they aren't really equivalent scenarios, I'm just wondering how much I actually know, and how much of it was me coasting through like a lot of these candidates
I am forever filled with impostor syndrome
my final words, as I lay on my deathbed, having been a professor for years and years with a bunch of papers to my name will be
"they should never have given me a PhD"
YOU HAVE NO POWER HERE, GILL-CREATURE!
anyway here's another interesting factoid
vanilla is expensive. Such that some producers will individually brand every bean to help track their shit to prevent theft
big image for detail:
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
this travesty of justice. . .
then read @Evil Multifarious 's [chat] OP
sorry mang
that'll be your thing
the state of our forum institutions remains strong despite recent events #blessed
NNID: Hakkekage
oh god no
I'm overall pretty pleased with how this all worked out.
I am torn
because
1) oh god no
2) that'd be kind of hilarious, honestly
3) oh god no
What kind of monster are you?