Moving from product data science to research science at meta
Is it possible/common to move from a product data science role at meta to a research science role? I have qualifications for both, but only ever get contacted for product data science roles. Curious if anyone has made the move or has a view on what that looks like. TC 200k
I heard that research scientists at Meta are title reserved for SWE with PhDs (easier for green card filing purposes). If you’re talking about people in FAIR, you would probably need PhDs any way or a lot of publications.
So I’m an applied scientist at amazon going as SWE-ml to meta. When I asked about title recruiter said everyone’s SWE but internally you can set whatever title you want in workday. Does this mean they won’t file my Gc as a scientist?
Should it not be the other way around: internally, everyone is an SDE, but the external title (what you say on LinkedIn) may be anything you prefer?
There has been cases of product ds moving to CDS.
Product data science at Meta isn’t just product analytics in other companies? It’s really not the hardcore side of DS
Most other companies these days also refer to it as product DS. Product analytics roles are getting rarer and more limited in scope. Google is one of the few that still calls it product analyst as opposed to product ds
Pinterest calls them Product Analysts as well. Not just Google
Which level ? Yoe ?
Is it a downlevel to move from product experience or product growth analyst teams to product DS? Or are they all at same levels?
Not anymore. Now it's a lateral transfer. Moving to SWE or PM is a downlevel though
Thank you! Is there a minimum tenure required before applying to transfer internally? I understand it's a full loop for interviews just like external.
These are quite helpful, thank you guys. A related question, I was asked for MLE interview from meta but i prefer more of research science positions. How these two differ? My understanding is as MLE you mostly build models and deploy whereas as RS you might more time doing research and methodology to solve a problem. Please let me know your thoughts
RS propose new models and MLE execute and maintain
Just to clarify you mean research scientist or research data scientist? Its confusing, but research data scientist roles are more like applied ML roles and dont have as high of a qualifications bar compared to research scientist (i.e. dont need a phd but many do have one) though they do a lot of cool stuff and have a very similar interview loop.
What cool stuff do research data scientists do? Is it not restricted to infrastructure problems like forecasting and capacity planning?
Cool is relative i suppose. I just meant cool in terms of real ML work. Infrastructure as a domain is understandably pretty dull to most.
DS is paid less, won’t beat ur current tc a lot
I saw both DS and ML for product analytics. Are they the same or different?
It's possible, I know people who have done similar things, I wouldn't recommend it. It's basically as much work as doing full eng interview loops elsewhere, and your TC will be significantly below your peers for an extended period of time after you switch since the initial grant is much lower for product DS. All but one data point I know who did this left FB within a year. Edit: oh, and forgot to mention, there's typically a down-level when transferring into SWE ladder roles (although not always).
Great insights. Thanks for sharing.
Why are product DS roles considered lower and given less grants?