HN user

dookahku

221 karma

data science, machine learning, C/C++, embedded, systems programming

Posts10
Comments105
View on HN

I'm wondering what's the objective?

I typed in a bit of an answer, there's not a whole lot to go on, there's no room for me to ask questions about this scenario or get data.

When I filled out an answer it just told me I was useless. I don't really get the point. What was I supposed to learn?

working hard at creating value does,

You'll have to explain to me what a person at work is doing, then? I mean, if they're not creating excess value by working at a job, whose value is then sold at a surplus over what they get paid, why does that job exist? Why did they get hired?

What exactly do you think people at 'normal jobs' are doing at work all day?

McNamara Fallacy 2 years ago

this reminds me of when Yahoo banned work from home in the late 2000s because they said the VPN data said they weren't working

how productive can a person bee if they wake up, immediately commute to work, labor for 12 hours(!), and then immediately commute home, sleep?

i've thought about that, trucking could be vulnerable to AI in the next few years...but after Starsky closed, I dunno. Lots of AV shops are kinda in doubt, but we'll see.

thanks for the advice.

i'd be happy with a regular software job, for sure.

I wouldn't say sustaining engineering (the bugfixing) was 'coasting' exactly. It's hard work, like a a crime drama every week, day. it can be quite challenging. :)

that's a good point...i have this fear because i've had terrible experience interviewing. i've been closer to the edge than this before...

one reason I stayed at my previous place for 10 years was because i just couldn't interview well...i know better about modern interview practices, which is why i practice so much ( 2-3 hours per day). it feels so far away tho.

i had the same thoughts, but one usecase was to quickly create GPU calculations. i wanted to allow users to create GPU kernels but without having to write a kernel.

Data has to get transferred from the GPU eventually, so breaking it up into steps back and forth does seem like over (under?) kill. Hopefully, since they're pipes in memory it won't be so bad as going to disk.

I'm surprised nobody is talking about a common understanding that IQ is not what separates the best startups from everyone else... I seem to remember PG saying that the initially thought that the smartest people would be the most successful in HNs early days but found out that's not the case