Hacker News Reader

In a very different writing style, but also mocking the technical interview: "Fizzbuzz in Tensorflow" (2016)

https://joelgrus.com/2016/05/23/fizz-buzz-in-tensorflow/

To whet your appetite:

> interviewer: Um, you understand the problem is fizzbuzz, right?

> me: Do I ever. So, now let's talk models. I'm thinking a simple multi-layer-perceptron with one hidden layer.

It would be very easy to turn that into the kind of story that is being shown here though. Just make the interviewee seem mystical or whatever and reply in terms like "now the machine must learn, please join me in the prayers needed to ensure the safety of the journey" and then have a giant paragraph of Buddhist sutra translated into Lojban or something.

This is actually a nice tutorial

I never fail to re-read the entire series when it's brought up and will do so right now.

I was going to link to this as it immediately rang similar in my mind!