How to hire engineering talent without the BS02 Mar 2023
Building a culture of inclusive technical interviews
“Team leads review candidate’s whiteboard, binary-tree inversion solution” from Classic Programmer Paintings
As a candidate in my early career, I experienced all sorts of technical interviews, ranging from being asked to write an algorithm on a piece of paper with an ink pen (not kidding) to doing leet coding in an IDE. Looking back, the best experiences were when the interviewer wanted me to succeed, was empathetic, and presented me with a problem that reflected how I would solve problems in the real world without gotchas.
Since then, I’ve been on both sides of the table hundreds of times and reflected a lot about the problem that plagues technical hiring: how to assess whether a candidate is both technically proficient and has the right mindsets and behaviors to succeed in your organization.
Most organizations have a technical interview process that attempts to evaluate both qualities to varying degrees of success. They may also use platforms like HackerRank or Karat to assess technical strength. But all too often, these approaches can lead to suboptimal outcomes, filtering out great talent or, worse, providing a poor candidate experience that hurts the company’s reputation.
We all know the head-scratching brain teaser, “Why are manhole covers round?” that was part of the Microsoft interview loop, which has become somewhat of an industry joke.
Also, we’ve seen some famous examples, such as the creator of Homebrew getting rejected by Google because he couldn’t solve a leet code problem. In fact, this is such a prevalent problem that engineers have started maintaining a list of companies that don’t have a broken hiring process.
Google: 90% of our engineers use the software you wrote (Homebrew), but you can’t invert a binary tree on a whiteboard so fuck off.— Max Howell (@mxcl) June 10, 2015
Most hiring managers, including myself, will admit it’s hard to nail technical interviewing if we are not super intentional about it. In my experience, I’ve observed several pitfalls we should all keep in mind if we want to build an inclusive hiring culture — or, dare I say, “fix” technical hiring.
With this post, I’m hoping to highlight some of those anti-patterns, or what a good technical interview is not, and then dive into some things that make for an exceptional interview.
So let’s first start with attributes that may seem like logical things to assess on the surface but aren’t representative of real-world experience and are likely to filter out great candidates
Simply recalling syntax, algorithms, concepts, or talking points doesn’t necessarily signal that someone will be a great engineer — it just shows they can think on their feet and recall facts and figures. Maybe they have spent a lot of time memorizing concepts without a deep understanding.
It’s always best to let the candidate look up information if it involves remembering specific details or, even better, having links to relevant references handy to share with them to save time.
Measuring the speed at which a candidate solves a problem doesn’t always correlate with how effective or efficient they’ll be in their day-to-day work. They may have just practiced a lot for coding challenges or interview questions. In fact, you may be doing a disservice to yourself by filtering out slow thinkers or neurodivergent candidates that are likely to not shine thinking on their feet.
Alright, so what makes for a good interview? In my experience, it’s essential to recognize that a successful interview is not just up to the candidate; the interviewer also plays an equally important role. Here are some attributes to get right in order to conduct an effective technical interview.
There is no advantage to creating a surprise factor. Give candidates a heads-up about the attributes or topics the interview will cover and any other information you can reasonably share upfront.
Consider using a candidate experience platform such as Guide, which makes it easy to set up an engaging candidate hub with all the information they need in one central place. This will go a long way in reducing anticipation anxiety and allow the candidate to show up confidently.
The interview should aim to assess the candidate’s deep understanding of concepts and skills that reflect what they will be doing in their day-to-day job. Rather than focusing on shallow technical concepts like a quick sort algorithm, ask questions that require a deep understanding of what the candidate will actually be working on, like building an API or a small service for example.
Some ways to make it even more realistic beyond the standard interview format:
Paid Trial — Offer the candidate a small paid project with you and your team. Of course, this is a very time-consuming process, so it’s more realistic when you don’t need to hire at a fast clip.
Take Home — Provide a take-home assessment option for candidates with a live component to discuss the solution they came up with to get signals on a deep understanding of what they’ve put together. The industry is split on take-homes, and there are several tradeoffs to the live whiteboard vs. the take-home, as laid out here by Andrew Rondeau but offering candidates a choice can help strike a balance.
Pair Programming — Another option would be to have one of the senior engineers on the team pair with the candidate on a problem where they would build on an existing project by adding a feature or two. This would give them a sense of what it would be like to collaborate with them while also closely mirroring how engineers solve problems in the real world.
A mountain of research-backed evidence shows that unscripted conversations are ineffective and lead to biases, interviewer idiosyncracies, and reduced hiring accuracy.
Ensure you have a set of attributes you look for and are aligned internally across the interview team to ensure a consistent candidate experience. For example, you could aim to get signals on dealing with ambiguity, code structure, testing, and user experience. You then want to ensure all interviewers know what great looks like for each. Ideally, you have a point-scoring system to determine how a candidate performed.
When you add new interviewers to the pool, ensure they are well-calibrated with the rest of the group by having them shadow and reverse-shadow interviews. After each interview, they should debrief with their counterparts to ensure they are aligned on the score. This will ensure interviewers’ subjectivity and personal biases are not creeping in.
Setting some expectations and guiding the candidate through the interview is important. Don’t expect them to read your mind in terms of what you are looking for. For example, I always tell the candidate upfront how I would like to spend our time by setting the agenda and expectations for each interview portion.
You know the attributes you need to get signals on. If the discussion is not going in a direction that will yield valuable signals, steer the candidate back on course.
Don’t hesitate to interrupt and ask if the candidate missed something. I promise you are not “giving away” the answer to the test. This is not an academic setting. In fact, this is an excellent opportunity to have a rich conversation and assess their self-awareness and communication skills.
Remember that even the most experienced candidates can feel nervous or unsure during an interview. Show some empathy and try to put the candidate at ease. Acknowledge the pressure they might feel and let them know there are no “gotchas” in the interview. Imagine how you would have felt in their shoes, and think about what would have improved the experience.
Don’t forget to get back to the candidate after the interview in a reasonable amount of time, ideally no more than 72 hours, to inform them about the decision. If you are not handling the comms with the candidate, ensure your internal team receives your post-interview feedback ASAP so they can keep the process moving.
Ultimately, it’s important to remember that technical hiring isn’t just about finding the most technically proficient candidate — it’s about finding the right fit for your team. Creating a realistic technical assessment can help you identify top talent and build a strong, diverse team ready to tackle any challenge.
And remember, it’s not just about what the candidates can offer you — they’re also evaluating your company and the hiring process. So, it’s crucial to create a positive candidate experience that shows them you value their time and contributions. By being empathetic and understanding during the interview process, you can make your company stand out as a great place to work.