> A HR professional can identify the candidates who don't come close to meeting the requirements of the job and filter them out.
In my experience this is mostly not true. Most HR people don't know tech so what the HR screen devolves into is cramming your CV with acronyms so you make it pass the HR grep filter.
> You will want to interview a decent number of folks for every position.
If you're bringing in more than five people for onsite interviews then (IMHO) you're doing it wrong. You're wasting your time and theirs and it's clear to me that you don't know what you're looking for.
At least Fred advocates phone/video screens (of 15 minutes) although he suggests 6-12 onsite candidates.
> Many employees don't know how to interview and you should teach them the basics ...
This is true but I would go further: most people shouldn't interview. It's arguably a talent not a skill. At Google, it's viewed as every engineer's responsibility to do interviews and I disagree with this. Many of the stories of bad interview experiences can be attributed--at least in part--to people who just don't have the interest or talent in interviewing.
Anecdote: in one quarter I once took 10 people to lunch as part of their interview slate. In those lunches I just talked about the company, answered their questions, etc but didn't get into any technical discussions.
Of those 10 I predicted 8 wouldn't get offers, 1 would and 1 I was borderline. Actual results: 1 offer, 9 no offers putting me at 9-10 out of 10.
> If you connect to the candidate on LinkedIn, you can quickly figure out who you know that knows them.
What an incestuous world the NY/SF tech scene must be if this is true. Well it might be true if you're a veteran VC. I can't see it being true for anyone else.
Otherwise this is all straightforward good advice although it skips the most important step: how to determine if someone is a good fit and competent but that's a topic we'll probably argue about in perpetuity.
1) I could tolerate HR screening for completely random people, who apply via a web form or cold email. Of course, I don't think I have ever hired or seen someone hired that way.
Anyone who comes in as a referral (from investors, employees, friends, etc.) shouldn't be speaking to anyone in HR until after an offer letter is signed. The only possible exception would be having HR handle travel arrangements for on-site interview, but even that should probably be handled by whoever does travel for the team in general (or the hiring manager if everyone normally does it themselves).
HR involvement will basically add zero value AND screen against your most competent 10% of candidates.
I'd probably walk out if an HR person talked to me, or someone tried to put HR in as a gatekeeper. I have met a couple of minimally competent HR people ever, and many more who were criminally incompetent.
2) 2-3 interviews with multiple people in each can work. It really depends on the role. What I'd probably prefer is 2 real interviews with 1-3 people in each, and then lunch where you get to meet a bunch more people casually (and basically be sold on the company), and then after lunch, an interview with founders (if they weren't in already). You can short circuit and send people home after the morning meetings if they're horrible, and after lunch if they're not-hire-now-but-maybe-in-the-future-or-may-have-competent-friends.
What an incestuous world the NY/SF tech scene must be if this is true. Well it might be true if you're a veteran VC. I can't see it being true for anyone else.
Tech is a pretty small world in general. These hubs may have more people overall, but they are also highly involved, highly connected people. A degree or two of separation is very, very common.
In my experience this is mostly not true. Most HR people don't know tech so what the HR screen devolves into is cramming your CV with acronyms so you make it pass the HR grep filter.
> You will want to interview a decent number of folks for every position.
If you're bringing in more than five people for onsite interviews then (IMHO) you're doing it wrong. You're wasting your time and theirs and it's clear to me that you don't know what you're looking for.
At least Fred advocates phone/video screens (of 15 minutes) although he suggests 6-12 onsite candidates.
> Many employees don't know how to interview and you should teach them the basics ...
This is true but I would go further: most people shouldn't interview. It's arguably a talent not a skill. At Google, it's viewed as every engineer's responsibility to do interviews and I disagree with this. Many of the stories of bad interview experiences can be attributed--at least in part--to people who just don't have the interest or talent in interviewing.
Anecdote: in one quarter I once took 10 people to lunch as part of their interview slate. In those lunches I just talked about the company, answered their questions, etc but didn't get into any technical discussions.
Of those 10 I predicted 8 wouldn't get offers, 1 would and 1 I was borderline. Actual results: 1 offer, 9 no offers putting me at 9-10 out of 10.
> If you connect to the candidate on LinkedIn, you can quickly figure out who you know that knows them.
What an incestuous world the NY/SF tech scene must be if this is true. Well it might be true if you're a veteran VC. I can't see it being true for anyone else.
Otherwise this is all straightforward good advice although it skips the most important step: how to determine if someone is a good fit and competent but that's a topic we'll probably argue about in perpetuity.