Ask HN: Taking a hiatus to learn more ML?

This sounds like a bad idea. You are very smart to have asked before proceeding.

I work as an MLE at a growth-stage startup with ~20 MLE/MLS folks. I was unemployed for 6 months before getting this job with 4YOE as an MLE and DS. There are too many qualified candidates.

ML research is out of the question. Most of the people who get to do ML research have PhDs, if not the only people. This seems like a racket but it exists for a good reason. It's hard for employers to evaluate the quality of MLS candidates through an interview, they don't know what to ask him. And if they've hired one, it's hard to know whether to fire him, things rarely pan out in research. The whole time, they have to trust this dweeb to run experiments burning tons of $$$ in compute! Employers are wise to be risk-averse, and to defer to costly social signals.

If you're going to take yourself off the job market for a long time, you had better at least get some kind of legible social signal out of it, like a master's degree. Almost all of the MLEs I work with have at least an MS in a relevant subject, the rest have PhDs.

I do have an MS in DS/ML, but thanks for the advice.

Now, if I were to look for ML engineering positions after that hiatus, would that change the answer? I already have some experience in that.

(I should have been clearer that I’m not expecting a research scientist job right away, I’d like to upskill and then take a job I’m qualified for, and try the PhD route later.)

That's extremely pertinent information, why didn't you say so?

An MS probably doesn't have that much more of a causal effect on your ability to do good ML work, as opposed to an equivalent amount of diligent self-directed study, which is what I assumed you were considering. But *the hiring market is dumb*, hence my spiel about social signals.

So now I'm less confident about the hiatus. I still don't think it's a good idea, but certainly not as terrible, and I think the HN consensus would be slightly in favor of it.

If you haven't already, consider posting this question to teamblind. Yes it's an incredibly toxic community, but it attracts people ruthlessly interested in maximizing their compensation. HN is biased towards entrepreneurship.

I didn’t think to say it because masters are basically undergrad and a requirement for engineer positions in my country (European 3-2-3 year system).

Thanks for the advice! I’ll think hard about whether I really want to do this.

Is it possible to do this with just a bachelor’s degree?

If you wanted to go back and do a PhD in ML, with your work experience you would be a great fit for the NSF Fellowship called CSGrad4US that supports engineers going back to research. https://www.nsf.gov/cise/CSGrad4US/ You have to be a US Citizen, and commit to a U.S. based university and in a CISE department (this is most CS departments). I am in the fellowship now and highly recommend it! Happy to answer any questions.

Not a US citizen but thanks for the info! Will check if something similar exists for my situation.

Single with no children or family commitments? If you really feel the pull, go for it, otherwise you'll be thinking "what if?" in a few years.

My thinking too. But maybe those savings aren’t enough to pay for compute while I’m practicing :)

Consider applying PhD program to some ML labs in Europe other than UK and France, they pay livable salaries like junior engineers. Better if in Swiss since the salaries are much higher.

European ML labs generally have less prestige than US/China ML labs but they pay much more humane wages even for people with family.

That would be nice. But I’m not from a great school and don’t have a lot of published research (co-authored a few papers where I wasn’t the main contributor). I thought maybe that would be a better choice in a few years with a better CV.

That's not necessarily a deterrent :) it really depends where, also would you consider staying in the states for it? Because I know of some reputable places (west coast mostly) where pretty much if you are to the right person's liking and fit their requirements, your admission won't see objections- there's plenty of time to make a decision about it, because the applications (iirc) take place a bit before* the end of the year. Let me know if you need more details!

I really should have mentioned that I’m not American, I just want to work there in the future.

Point taken about selection though! I’m thinking I could network with academics while working on implementing papers, etc.

Do you think the selection situation is different for foreigners? Do you know if the same is true about European schools? The 3-year average for PhDs in Europe really looks more attractive to me.

Also would really appreciate if you could tell me more about how these programs select people who aren’t 4.0 students from a top school with famous recommendations. Is practice important, should you know a lot about current SOtA, do you need to show math proficiency?

Good thinking, and regarding Euro institutions there's lot of regional differences- for instance, brits have steps to take which are pretty set in stone with the Centres for Doctoral Training; basically it's the opposite of the process from the prior comment, it's the system that allocates someone for you. Denmark is free if you're native or from EU- Finland however is free for all (according to peers and the net) and would gauge they're both tad more flexible than UK in terms of requirements. Anyway, there's plenty options! Regarding the selection process, although they're a clear benefit perfect grades aren't a must and it's alright to be from outside of a top ranking school- but you'll have to bring something to the table to make up for it. Whether that is a solid paper, practical achievements along recommendations or whatever again depends on the place. I think there's just so many ways to show something that can be impressive without knowledge about the latest thing or dropping walls of equations for the sake of it haha. Best of luck!