The documentation for each interview should:
- Define what the interviewer should be trying to measure and what "good" looks like.
- Provide questions that the interviewer can ask to get a good signal on the things that they need to measure.
Interviewers are not required to linearly ask every single question in the interview documentation; it is not a script.
- If an interviewer gets the desired signal after asking only one our of three questions on a topic, then they may decide to skip the remaining two questions and move on to another topic in the interview.
- If the candidate brings up an experience or topic that is relevant to another section of the interview, the interviewer may choose to use that as a natual segue to that topic even if it is out of order from the interview documentation.
The best kind of interview is one where the candidate feels like it was an organic conversation and the interviewer got all the signal they were looking for.
We would like to move as much information as possbile about our interviews to our public handbook, including the interview questions themselves. This will allow motivated candidates to prepare their best answers to our questions. This work will be done by, and at the discretion of, individual hiring managers for the processes that they own.
For each interview we ideally want:
- At least one member from the team that the candidate will be joining.
- Two interviewers total, to maximize the number of interactions between the candidate and our team, and to continuously calibrate our interview process.
Interviewers may be adjusted due to scheduling difficulties (e.g. interviewer availability, timezones).
We want to mimimize the bias in our hiring process, so we will try to avoid scheduling you as an interviewer for candidates that you have an existing relationship with (e.g. close friends, family, significant others, candidate who you refer). This may be unavoidable if you are the hiring manager.
We are flexible when it comes to scheduling interviews because we are all-remote and we don't need to schedule all interviews back-to-back (unlike typical onsite interviews at other companies). Some candidates like spreading interviews out (e.g., across multiple days and/or having breaks between interviews on the same day) and others prefer to batch them as much as possible to get it over with. Ask the candidate what their ideal interview schedule looks like and then try to accommodate those preferences as much as possible (given interviewer availability).
Even if candidates prefer to batch all their interviews together, it might not be possible due to the timezones of available interviewers.
If interviews are spread over multiple days and it is clear from the initial feedback that we won't be moving forward, we will cancel the remaining scheduled interviews to save ourselves and the candidate time.
Please put your written feedback into Lever as soon as possible following the interview (immediately is ideal) while it is still fresh in your mind. To reduce bias and herding, please refrain from discussing your impressions with anyone else on the interview panel until everyone has submitted their written feedback.
Your feedback should contain two parts:
- Objective summary of the facts of the interview (questions and answers) in chronological order.
- Your subjective assessment of how well the candidate responded to your questions.
You can interleave these together, or leave the assessment for the end, whichever you prefer.
- Strong positive. You are confident that this person would make our team better and you would be excited to have the chance to work with them.
- Positive. You observed only positive signs that this person would be a good fit for our team, but you don't feel strongly about hiring them.
- Negative. You are neutral or have doubts about whether this person would make our team better.
- Strong negative. You are very confident that we should not hire this person because they would not be successful on our team.
All positive feedback with no strongly positive feedback is usually not sufficient. We want to hire people who we are excited to work with.
We conduct debriefs after a candidate finishes interviews. The debrief should be added to the calendar at the same time we invite the candidate to interviews.
Should the hiring manager decide to not utilize the debrief (e.g. the feedback was overwhelmingly positive or negative), the hiring manager should post to the candidate's Lever profile stating the reason and/or context for the decision. This provides a sense of closure for those who interviewed the candidate.
Once interviews are over, the hiring manager may choose to cancel a debrief if the submitted feedback makes it obvious that we won't be moving forward with an offer AND if the outcome doesn't raise any questions from the hiring manager about the interview panels or hiring process (e.g., was there misalignment about the role the candidate was interviewing for, was there conflicting feedback worth discussing, could we have filtered out this candidate earlier in the process).
The debrief calendar invite will include a link to a Google Doc that contains all the feedback (due to a limitation in Lever permissions). Everyone should pre-read all of the feedback before the debrief begins. The hiring manager drives the debrief discussion by asking specific questions that they prepared in advance (e.g, "I see you flagged a concern about X. Did reading feedback from other interviewers address your concern?").
| Sentiment | Outcome |
|---|---|
| Not enough excitement or serious doubts. | End interview process by not making an offer. |
| Interest in moving forward, but we are missing some signal. | Move forward with a followup call where the hiring manager asks direct and specific questions to gather the missing signal. |
| Sufficient excitement and no doubts. | Move forward with an offer. |
Candidate feedback will not be shared with anyone that didn't participate in the interview panel.