3 ms·
> you might find that you are not looking for a Data Scientist at all, perhaps you are in fact looking for a Data Engineer or a Data Ops person Strong agree M
by awinter-py 7y ago
> you might find that you are not looking for a Data Scientist at all, perhaps you are in fact looking for a Data Engineer or a Data Ops person
Strong agree
Managers who haven't been through this process once will hire a smart stats person who can't operate without a full-time engineer pulling production data.
Which can be crippling for small teams. The alternative is to make your PM do their own data work or promote a product-minded engineer into this role.
- bradleyjg 7y agoYour PM can make you good looking charts and do basic summary statistics (mean, median, percentiles, etc.) I would call that business analyst work. It's probably good enough in a lot of situations. But if you actually need statistical analysis, there's no substitute for a statistician. Even working scientists in the hard sciences routinely screw up their stats and should be (but mostly aren't) consulting with a staff statistician for their papers.