Why You Should Design a Better Hiring Process
24 September, 2021
Getting engineering interviews right is a tough problem. As with engineering, the types of interviews we run come with a set of tradeoffs. Early on in Policygenius’ life, we interviewed senior engineer candidates with a different process than we did for early- and mid-level engineers. For engineers with less experience, we would have take-home assignments; so before coming on-site, they would spend an hour or two workings on a concrete problem that we gave them. If they passed the first round, we would work with them on expanding what they had developed in a series of interviews. For Senior candidates, we didn’t use this model, because senior candidates were less willing to do these types of assignments.
While we got some good signals from the take-home assignments, experienced candidates have a lot more choices. To some extent, they have more control over the job marketplace.
There is no one consistent definition of senior engineer across the industry. In a lot of cases, when people were on the borderline, they would be sitting through the wrong series of interviews if they were more senior or junior than we expected. Transitioning people between these two interview pipelines was impossible due to the take-home pre-work.
Our challenge was creating an interview process that created consistency in our process, but also allowed us to reliably determine the level for different candidates, regardless of their title at other organizations.
We had some big challenges ahead of us in creating interview funnels that worked for all parties involved. How could we design interviews that successfully capture a wide range of candidate abilities, in a way that’s objective across all of the interviewers in our organization?
We turned to the root of those determinations: our career rubrics, which we use to define each of our roles and the expected competencies of engineers at various levels of their careers. Which of those did it feel we could reliably assess in an interview? Which of those couldn’t meet that standard?
We took down that list and identified which of our varied existing interview modules could map to each of those competencies. By evaluating these for each existing interview module, we could determine which interviews were overlapping, which were critical and trim out the components we didn’t need.
Using those same competencies, we built out a new form of rubric – rubrics that evaluated each and every one of our interviews, and identified how we expected candidates of different levels to perform in those interviews. These weren’t checkboxes but were guidelines for interviewers to use to help them objectively assess. Interviewers loved these new rubrics – it gave them more confidence in their interviewing and gave them a foundation on to build their interview feedback on.
We also went to the drawing board and created all the interview questions that captured this wider range of ability a lot better than in the past. There was no perfect translation of their skill sets, but we did give our best to build interviews that gave us the flexibility to evaluate many potential areas. Using those rubrics as a foundation, we could carefully design questions to capture exactly what we wanted them to evaluate.
No interview scenario is perfect. Different candidates have different wants from an interview, and you can’t objectively meet them all where they want to be met. By creating an interview structure and unifying our pipeline, though, we found that candidate experience improved, recruiters and coordinators were able to easily adapt to different candidates, and hiring managers got better signals than ever before regarding candidate experience.
- Interview rubrics are incredibly important. Many different organizations did not know what they were looking for, but the rubrics score all the candidates' responses against the same criteria. It eases the evaluation process.
- Build an interview process that doesn’t make too many assumptions about a candidate’s experience – they may well have a lot more to offer than it looks like on paper!
Scale your coaching effort for your engineering and product teams
Develop yourself to become a stronger engineering / product leader
Shubhro Roy, Engineering Manager at Box, takes nothing for granted when refining the way that he and his team evaluate the work and line it up for production.
Engineering Manager at box
Shubhro Roy, Engineering Manager at Box, establishes a steady rhythm of mentorship by planning bi-weekly one-on-ones with his direct reports.
Engineering Manager at box
Mustafa Furinturewala, VP of Engineering at Coursera, speaks on overcoming the constraints of hiring in the Bay Area by opening a Toronto office and emphasizing the importance of local recruiters, brand awareness, and team culture.
VP Of Engineering at Coursera
Maulik Bengali, founder and CTO of Ajackus, shares how he scaled his technology team from 18 employees to 80 within 18 months.
CTO at Ajackus
Maulik Bengali, founder and CTO of Ajackus, speaks on how he evolved as a founder & CTO by improving his collaboration, preparation, and core engineering skills.
CTO at Ajackus
You're a great engineer.
Become a great engineering leader.
Plato (platohq.com) is the world's biggest mentorship platform for engineering managers & product managers. We've curated a community of mentors who are the tech industry's best engineering & product leaders from companies like Facebook, Lyft, Slack, Airbnb, Gusto, and more.