There seems to be a lot of criticism of the suggestions here, but I think they are good ones. It seems to pretty much boil down to:
(1) Spend a lot of time discussing accomplishments at their current / previous employers. If you drill down on particular projects they should be able to explain, in great detail, exactly what their contribution was and how they worked with the rest of the team to get that done.
(2) Pick a technical question that has a lot of depth so that you avoid false positives because people have just looked up the answer on the internet. I would also add that in my view the best questions have no tricks and require no "aha!" moments. It should be something the candidate can be reasoned through step by step so you can see their thought process. When they come up with an initial solution, then you can dive deeper by expanding the problem or adding additional constraints.
I'm also disappointed that so many comments above do not engage with the information in the article. I, too, found this to be a thoughtful and helpful piece. Along the parts I liked --
Using "calibrated" questions that you have used before and know how far people get with.
Focusing on a specific accomplishment or area of strength of the candidate to probe. (As opposed to choosing something from the resume that, say, the interviewer was most comfortable with, or that the interviewer found interesting.)
Paying attention to preparing the interviewers, so they know why the candidate is being considered, they know why they are on the interview list, and they are experienced enough with interviewing to give a useful opinion.
(1) Spend a lot of time discussing accomplishments at their current / previous employers. If you drill down on particular projects they should be able to explain, in great detail, exactly what their contribution was and how they worked with the rest of the team to get that done.
(2) Pick a technical question that has a lot of depth so that you avoid false positives because people have just looked up the answer on the internet. I would also add that in my view the best questions have no tricks and require no "aha!" moments. It should be something the candidate can be reasoned through step by step so you can see their thought process. When they come up with an initial solution, then you can dive deeper by expanding the problem or adding additional constraints.