Part 2: How To Not Lose Female Engineering Candidates

Like I said yesterday, I’m thankful for my smart friends. After Paradigm’s Joelle Emerson provided some feedback on how Take-Home Code Projects might impact gender ratios in hiring pipelines, another amazing woman sent me her perspectives. Kieran Synder is CEO of Textio, an innovative Seattle startup which helps companies understand the language of their job listings. Kieran’s email, which she’s letting me publish here:

“We do a take-home project too, longer than 3–4 hours actually. We have not had women pass at a higher or lower rate than men, nor have we had anyone other than one man decide not to move on in the process after seeing the question. We’ve been doing this for three years — and we do take-home projects of similar character for every role (obviously, for non-engineer roles, the project is something other than coding).

We love this way of interviewing for a whole bunch of reasons.

We think it’s fairer than a whiteboard interview, because it establishes shared context for 1–1 interviews. It lets people work like real developers work, using Google or Stack Overflow or whatever other resources they want. Both people we’ve hired and people we haven’t hired have given us good feedback about it. Many, many people have told us after their interviews that the main reason they come away excited to join Textio is for the chance to work with the people who interviewed them.

I will say there’s an art to writing the questions so they are both constrained and open-ended enough. And it’s important that the group session where people present their work not feel like a firing squad. If there’s anything I wish we could fix, it’s the instant anxiety that some candidates feel walking into a room that includes Jensen and/or me.

The other thing I’d say is that teams telegraph a lot even during initial screening, before the longer take-home is ever introduced. We have a reasonably diverse team and make sure our screening and interview processes reflect this. If women have a bad experience in the screening, they might choose not to move on regardless of the definition of the follow-up task.”

Thanks Kieran!