I wouldn't put too much weight into this. Not only is much of this content dated, but Chip is far from a subject matter expert. She loves to write (and is a greater writer), but don't expect anything beyond a cursory introduction.
The Deep Learning Interviews book (more specifically volume 2, based on the proposed contents) in the other thread is much more representative of ML interviews for candidates with at least an undergraduate level of machine learning training.
Note machine learning engineering is very different from model and data work, i.e. designing the experiments. There are plenty of jobs where you package Nvidia drivers and pytorch files into docker containers, or write low level C++ to e.g. implement a transformer network on a new device architecture. Those require nothing more than a cursory knowledge of machine learning, and you can essentially get away treating them as magical black box matrix multiplication formulas. Very few companies can actually afford the 7 figure salaries for actual frontier level machine learning research.
For example, if you want to run a GPT model on some obscure graphics chip, you are better off hiring a C++ computer graphics/embedded engineer to do it than a typical academic trained ML researcher. The engineer can implement a GPT model simply by building out the matrix multiplications, and can do a better job without even knowing what an activation function is.
I did a cursory browse through on few sections of this current book (namely the CV module), and I think the questions are on the easier end for actual ML interviews/whiteboarding. Normally, I would face some more depth (and equivalently as a tech lead, similarly ask more than surface-level questions to potential hires).
tldr: If you have gone through an introductory ML course like Andrew Ng's CS229 or CS230, these question banks seem obvious & trivial to solve.
I think a question on a lot of people's lips is "if we put the effort in to retrain and use this guide, how much can I expect to earn?" All well and good saying this is what employers want in an interview, but we have to talk remuneration too :)
In the bottom left of the compensation graph, there is a non-zero percentage of people earning less than 0 USD per year. I guess they used a smoothing function fit to the data points. The alternative is that some people are paying their employers some dollars for the privilege of working for them :p
I wouldn't put too much weight into this. Not only is much of this content dated, but Chip is far from a subject matter expert. She loves to write (and is a greater writer), but don't expect anything beyond a cursory introduction.
I had a look, and being an interviewer for ML related positions, it’s a mistake to be using boilerplate questions like this.
Invest time creating your own pet projects rather than cutting corners with these books.
Are there other, more up-to-date, resources you would recommend?
The Deep Learning Interviews book (more specifically volume 2, based on the proposed contents) in the other thread is much more representative of ML interviews for candidates with at least an undergraduate level of machine learning training.
https://news.ycombinator.com/item?id=41084834
Note machine learning engineering is very different from model and data work, i.e. designing the experiments. There are plenty of jobs where you package Nvidia drivers and pytorch files into docker containers, or write low level C++ to e.g. implement a transformer network on a new device architecture. Those require nothing more than a cursory knowledge of machine learning, and you can essentially get away treating them as magical black box matrix multiplication formulas. Very few companies can actually afford the 7 figure salaries for actual frontier level machine learning research.
For example, if you want to run a GPT model on some obscure graphics chip, you are better off hiring a C++ computer graphics/embedded engineer to do it than a typical academic trained ML researcher. The engineer can implement a GPT model simply by building out the matrix multiplications, and can do a better job without even knowing what an activation function is.
I found the "Deep learning interviews" book to be much more engaging and value for time.
https://arxiv.org/abs/2201.00650
I did a cursory browse through on few sections of this current book (namely the CV module), and I think the questions are on the easier end for actual ML interviews/whiteboarding. Normally, I would face some more depth (and equivalently as a tech lead, similarly ask more than surface-level questions to potential hires).
tldr: If you have gone through an introductory ML course like Andrew Ng's CS229 or CS230, these question banks seem obvious & trivial to solve.
I think a question on a lot of people's lips is "if we put the effort in to retrain and use this guide, how much can I expect to earn?" All well and good saying this is what employers want in an interview, but we have to talk remuneration too :)
"RTFM"[0] as they say
[0]https://huyenchip.com/ml-interviews-book/contents/3.1.1-base...
In the bottom left of the compensation graph, there is a non-zero percentage of people earning less than 0 USD per year. I guess they used a smoothing function fit to the data points. The alternative is that some people are paying their employers some dollars for the privilege of working for them :p
There are some weird waves at 400k, 500k, 600k, ... It's probably an histogram diguised as a line graphic.