Comment by caffeine | Hacker News Reader

Comment by caffeine | Hacker News Reader

I think it's helpful to be more specific about the role, since "data science" is a vague term these days.

If you are using existing methods to analyze data relevant to your employer (so like trying to optimize the conversion funnel at a startup or something), PhD is unnecessary.

If you are trying to invent new statistical techniques (say deep learning researcher at DeepMind), PhD is helpful.

Also, different groups have different cultures. Some groups want to function like academia, with reading groups and paper talks and stuff. Some want to function like software companies with standups and Trello boards. A PhD will fit in better with the former type of group.

Edit: I dropped out of a PhD to go work, and that was optimal for me. I got a couple extra years to independently build my skillset, I was very hot on the recruiting market being a step above the new grads, but I didn't waste ages learning skills that would ultimately be irrelevant (published no papers.)

Just adding to your point to define what exactly data science is these days. I've been interviewing intermittently for the last few weeks and different companies mean different things when they mean data scientist. It's a catchall-term which includes some of the following: data analyst, data engineer, business analyst, machine learning engineer, product/growth analyst etc. For all of these, I would say that a PhD is not required.

Another thing I've observed is that even when a PhD is required, companies usually add "or equivalent work experience" (4+ YOE) in their requirements.

Thanks for the response. I already work as a data scientist (no PhD, big company, a couple of years experience) and I'm much closer to the no-original-research software-team type of work. My thinking has been similar to what you describe but I'm still having some doubts about missing out on the PhD route.

I see. Knowing now where you are coming from, I think that there are headwinds against your progression in an elite ($500k/yr +) direction without a PhD, but no hard barriers.

That can be mitigated if you have the mathematical / statistical chops, or if you happen to work at the right place on the right problem.

So basically a PhD would be momentum for you to get on the elite track, but there are still plenty of ways to stand out. If you are the guy who designs the research infrastructure, or builds the team, and the group is wildly successful, you can still make it.

> since "data science" is a vague term these days.

Many don't think it should even be a term at all.

Reply on news.ycombinator.com