A Complete Guide to the Amazon Bar Raiser Interview Process

Welcome to the Scarlet Ink newsletter. I’m Dave Anderson, an ex-Amazon Tech Director and GM. Each week I write a newsletter article on tech industry careers and tactical leadership advice.

Free members can read some amount of each article, while paid members can read the full article. For some, part of the article is plenty! But if you’d like to read more, I’d love you to consider becoming a paid member!

We are honestly terrible at picking which people will be great co-workers. We review resumes and search for relevant experience while combing through overstated or poorly explained projects. We attempt to test, over a period of a few hours, technical competence that was built over years in complex environments. We ask for references, and pretend that feedback from candidate named sources is even remotely unbiased.

While we need to assess a candidate’s ability to functionally do the job, we also need to make certain they have the right leadership qualities to work with our teams.

Me, meeting with one of the engineers on my team.

Me - “So what's the main reason you'd like to transfer teams?”

Employee - “I can't work with Ralph anymore. He’s constantly screwing up. He never admits that it’s his fault when his code has a problem. He points fingers at any of us that make the slightest error. He fights over every single of his code reviews. He’s just making life on the team miserable.”

Employees need to be able to get along with co-workers, graciously admit mistakes, give credit when it's due, and at the very least, not make the work environment worse through their presence.

And while we want to filter for technical competence and positive leadership skills, we need to filter for those traits without mistakenly bringing personal bias into the process. Beyond illegal bias (e.g. gender, or race), we also need to be cautious about our biases regarding education, grammar, the types of companies someone has worked at (ewww, a consulting company?), or someone's choice to spend the previous year surfing instead of working. We are all biased in a thousand ways, and it’s challenging to identify which of our biases are valuable for filtering, and which aren’t.

This is all to say that interviewing is hard. It takes dozens of interviews to feel remotely competent. It takes hundreds of interviews to begin to feel some aspects of mastery. And that’s the crux of the matter. Amazon invented a bar raiser position to bring mastery and experience into every interview loop.

What is a Bar Raiser?

A bar raiser is an experienced person added to an interview loop to ensure that it does not go horribly off the rails.

Stated more optimistically, a bar raiser is responsible for ensuring that a candidate has a great experience, and Amazon makes a great hiring decision.

The bar raiser is an experienced leader. Their common sense and judgment are trusted by their leadership team, and they have the responsibility (and authority) to correct or modify elements of an interview process. Rather than a checklist of responsibilities, a bar raiser impacts the entire process from end to end. It is not a clearly defined role; rather, it's a leadership position that requires an understanding of what ownership means.

I’d personally say that this is the hardest aspect of being a good bar raiser. You have a vague charter, piles of unruly peers, and the responsibility and authority to do the right thing.

Pre-bar raiser involvement.

Before a bar raiser is involved with a candidate, a few steps have taken place.

A resume is first screened. This means someone (a recruiter, or manager, engineer, etc) looks at the candidate. This makes sense, because you wouldn’t want to do a screening interview without knowing a candidate's background. I often found that the most updated and detailed description of someone’s work history was found on LinkedIn rather than on their resume.

After a candidate has been flagged as interesting, someone conducts a screen interview. The primary purpose of this screen is to validate that a full interview should take place. This is for the benefit of the candidate as well as the interviewers.

I went through something like this more than once:

Recruiter - “Hey, your interview is next, right Dave? You're the third interviewer in the loop?”

Me - “Yeah, why?”

Recruiter - “Something went wrong with the initial screen. The candidate did terribly in the first interview, and are bombing the second interview right now. The second interviewer says the candidate has started to cry, even though they’re trying to ask easy questions.”

Me - “Ok, I'll talk to the candidate and see if they'd like to leave early.”

No one wants a candidate to have a terrible time. As a candidate, it’s one thing to make a mistake or two while interviewing. We’ve all said stupid things in interviews we’ve done, and I've received dozens of emails from candidates after interviews explaining that they have a much better answer now.

However, it’s a different ballgame entirely if you’re interviewing for a position you're not remotely qualified for. This can be a terribly embarrassing experience. The screen, almost always done remotely, is a way of ensuring that the candidate will be prepared.

One way of phrasing the purpose of screening is that our goal is to ensure a candidate gets at least one “yes” vote during their full loop. If we believe they have a shot of being hired and would likely pass one or more in-house interviews, that should be considered a pass in a screen. A screen is not intended to indicate "I think we should hire this person", it's instead a "I think there's a chance this person could pass an interview" decision.

I state this partially because I think it’s interesting, and partially because even internal Amazon interviewers get confused about this fact...