For a growing company, interviewing is one of the most important tasks employees are asked to do, but one challenge is that interviewers don’t always feel adequately qualified to assess candidates. Lacking shared standards, it’s easy to rely on gut feeling and subjective impressions, which can depend on mood and incorporate many forms of potential bias. Many agree that the technical interview process is broken.

At Medium, we’ve long recognised the need to be intentional in our hiring process and have been thinking about how to improve it for most of the company’s existence. At the beginning of the year, we decided to put renewed effort into improving Medium’s process for hiring engineers, updating our existing selection criteria and creating a rubric to help provide consistency across interviewers. While we have never based our hiring decisions on certain potentially misleading criteria like college name or GPA, we took the opportunity to make factors like this explicit and spell out what we do and don’t care about. This is important to ensure we maintain these principles as we grow rapidly.

We’ve been using these guides for the last few months and are excited about the results we’ve seen so far. The eng team did a day-long off-site in April to understand the rationale, get familiar with the material, and develop interview questions. We continue to provide ongoing coaching to our interviewers, and iterate on the rubric itself.

Today, we’re excited to share our hiring documentation for everyone to see. We are releasing four documents we put together under a Creative Commons Attribution Share-Alike license:

• The rationale for refining our process
• What we look for in candidates
• What we don’t look for in candidates
• How we grade candidates

To ensure our own accountability, the public rubric and rationale will be the actual guides that we use to judge candidates; as of today, there is no different internal copy.

Following the lead of our legal team’s transparency around our terms of service, we plan to put the rubric on Github so people can see how it evolves over time.

Many were involved in this effort. Thanks are due to Jesse Toth and

at Github, and

Karla L. Monterroso

of CODE2040, who reviewed early drafts. We were influenced in part by the writing of

Camille Fournier

, and inspired to release it in part by Clef’s handbook and Automattic’s legal documents.

CODE2040

convened a workshop where we got to hear some really smart perspectives from industry leaders, which helped shape the rubric. And internally, thanks are due to the many Medium engineers who offered thoughtful critiques and suggestions, including

Joy Chen

,

Jonathan Fuchs

,

Emma Zhou

(having been hired using it!), and

Kelly Ellis

.

Releasing this publicly is unusual and maybe unique. It’s also a little discomforting, but we believe it’s an important step that demonstrates our commitment to hiring objectively, based on the things that we think actually matter to job performance. We don’t claim to have all the answers, nor that this specific set of skills will be the ones that meet other organizations’ needs, but we hope that by sharing the documents under an open license, others can build upon our work in ways that works for them. We hope it sparks a discussion within the wider tech community about hiring best practices and look forward to learning from our colleagues in the industry about areas where we can improve; we expect our process to evolve continually.

Regardless, we’re interested to hear what people think. And if Medium sounds like a place you want to work, come and join us, we’re hiring!