The job description for Research Analyst says that the best candidates will have “comfort thinking in terms of expected value and using systematic, quantitative frameworks.” How quantitative should a candidate be to apply? For example, if a person feels comfortable with basic expected value concepts but finds GiveWell’s CEA overwhelming (and probably could never produce something similar to GiveWell’s CEA), is that not quantitative enough?
itty
Karma: 20
Who is likely to manage new Grants Associates, Operations Associates, and Research Analysts?
GiveWell is also hiring for several roles right now. How should a person decide whether to apply to GiveWell or Open Phil? Are there significant differences in the work culture? Do Research Analysts at each org take on similar types of tasks?
What are things that previous Research Analysts have struggled with at Open Phil? What are reasons others have found it not to be a good fit?
I imagine it’s complicated to release details about projects that aren’t selected to receive a grant. Presumably there are reasons the project wasn’t selected. If EA Grants wanted to publicize their rejects, seems like they have two main options:
1) Make the project public without explaining their reasons for rejecting it. In this case, EA Grants might be making it more likely that a bad project is funded, by bringing it to the attention of other funders without warning them of possible pitfalls.
2) Make the project public AND explain their reasons. This might work, but takes up more time and social capital. EA Grants would have to spend time figuring out how to phrase their reasons diplomatically, so that a) their rejects weren’t too discouraged or angry, b) other excellent candidates aren’t discouraged from applying, c) other funders are appropriately warned, etc. It’s hard to balance all of this. CEA also wouldn’t be able to disclose confidential information that their decision might rely on, making it even trickier.
In summary: I think this is a lot harder than it might initially appear. (That said, there might still be good ways to do it?)