On Uneducated Hiring

Hiring in the technology field is famously bizarre. Especially in larger companies the practice of a multi-round process that focuses more on “interview skills” than what the job is about just feels completely out of place.

When I was transitioning from a company where I was able to influence the hiring design1 into a company with a stereotypical Silicon Valley process2 that I had to represent3, I started to have a pet theory on why the hell is this happening.

I blame the education system.

The Structure of an Education System

In most countries the education system is a structured system designed to evaluate who should have a better shot at upward social mobility4. Through a set of exams it creates a funnel through which people go through to deserve a title that indicates their achievement.

While this system no longer holds in some countries within some occupations (technology being one), it is true for the majority of the population and a hard requirement for all certified fields (like healthcare or actual engineering). The importance of individual tests varies heavily from country to country, but SAT or gaokao verge on national psychosis.

This reaction makes sense. Our work shapes our identity significantly. You hear “I am a doctor” way more often than “I work as a doctor”. This means those tests are a one-time attempt at being who you want to be.

For a decade or two in our lives, we are conditioned to perceive tests as an evaluation of our worth.

What Is Hiring

I think the “test as a measure of an individual’s worth” spilled into hiring as it originally shared a similar purpose. As employees were expected to stay in a company for decades they were essentially admitted into a post-university. Corporations could have been sorted into tiers similarly to universities—based on how much they have paid, while students were allowed an entrance using a similar evaluation mechanism.

This is no longer true—especially in technology, and especially in the Valley. People usually stay a few years even at posh corporations. Hence, I propose a different definition of hiring:

Hiring is an activity designed to evaluate a fit between a candidate with their current skillset and a company with their current needs with consideration for future development.

The “evaluation of a fit” goes both ways: this is a conversation between two adult parties. Depending on the current market circumstances one side has more power. The way this power is (ab)used during the interview process also tells you something5.

If you are rejected from an interview, it’s not because you are not worth a job, it’s because there currently6 isn’t a fit between you and the company.

Unlearning The Process

Switching the minds of the interviewers from “evaluating a person” to “evaluating a fit” has been, in my experience, the most important starting point towards having the interviews work better7.

One important part is to untangle the power dynamic. Interviews can easily devolve into university-style “answering trivia questions in front of a panel of judges, trying to guess what is the answer the interviewer has in mind”. You could tell which interviews are like this just by watching the nonverbal communication.

This is also ingrained in candidates—I was able to summon life-threatening distress in candidates by asking “what do you like to do?”. It often took quite a bit of time to convince them that I am not asking them to guess what I think is the right answer.

The second important part is to structure the interview around the content of the work. A lot of interviews are structured about the hypotheticals, under the idea that “if you can do this, easier things will just be easier”. I don’t think it’s true. And since top engineers say writing top conference research papers is easier than interviews, I think I am not alone.

This also goes for culture. At larger companies candidates are evaluated against the stated values, but it’s borderline impossible to have completely consistent values in all teams—which gives room for an expectation mismatch. It also significantly limits the candidate pool as the company is hiring for culture fit instead of culture add.

I think there used to be some merit to “irrelevant” testing at larger companies because of large internal mobility. Once hired at a defined level, employees are expected to perform at that level anywhere in the company. But from what I observed in practice is that most teams do the same interviews for internal and external candidates—as, at scale, the inconsistency is unavoidable. And I think it’s not realistic to expect engineers to be equally senior in the frontend, backend, infrastructure, and machine learning, whereas a lot of corporations keep using the same job codes for all of those.


The collective trauma of the experience with the education system skews the design of our hiring. This affects the ability to hire well since interview skills (e.g. performance under focused group supervision, public speaking skills, and negotiation) are rewarded over job skills.

I think we should work more consciously on unlearning this and thus finding better ways to select people.

I’ve received feedback saying “this is nice, so what shall we do to improve hiring?”. I don’t feel qualified to answer that question (and I am happy to add links if you know some. I can only share my personal attempts, but that’s a different article.

Thanks to Daria Grudzien for proofreading and feedback.

  1. “How I Hire” will be another article ↩︎

  2. Broadly summarised as 5+ rounds, live coding and whiteboard coding with a focus on code correctness, knowledge of algorithms and algorithmic complexity, testing for compatibility or knowledge of specific company value claims ↩︎

  3. When I was finally overruled and forced to fit into the process, I was almost unable to hide my cringing in front of the candidates. It was a major reason why I left. ↩︎

  4. Definition obviously mine. While editing this, I was curious how experts define it and wasn't very successful at finding anything that's succinct. In my own country the goals of education are specified on pages 13-15 of the education whitepaper ↩︎

  5. Note that in case of power balance being on the employer's side, unless it’s the founder interviewing you, the power has been temporarily lent by the business to the person conducting the interview. ↩︎

  6. People do learn and change, as do companies. There is no reason not to re-evaluate after some time. ↩︎

  7. Defined as “higher retention of hired candidates” and “being able to hire more people”. “We have hired less than a percent of candidates” is often published by companies as a way to show pride in the quality of hires. In my experience this is just elitism, and I believe it related to the education system as well. There would be some merit to it—if everybody was completely confident that their hiring process selects only the people that are best at the job. I don't think such a process has been invented yet. ↩︎

Published in Essays and tagged

All texts written by . I'd love to hear your feedback. If you've liked this, you may want to subscribe for my monthly newsletter, RSS , or Mastodon. You can always return to home page or read about the site and its privacy handling.