Tech Interview Prep: How Scoring Really Works

This article is crossposted from IEEE Spectrum’s careers newsletter. Sign up now to get insider tips, expert advice, and practical strategies, written in partnership with tech career development company Parsity and delivered to your inbox for free!

I’ve sat on both sides of the interview table several times over the past decade. You might be surprised to hear that I’ve often been just as nervous interviewing candidates as I was when being interviewed!

Nearly all the interview advice out there is about the candidate’s side, but understanding the other side can also help you prepare. Let me show you what I’ve seen firsthand, and what I’d bet is happening at the company you just interviewed with.

If you recently got rejected after an interview, this might explain what actually happened.

One caveat, because I’ve been on the receiving end of this: A couple of my recent interviews were run entirely by AI. These were screening rounds, but a growing share of job seekers now report being interviewed by a bot somewhere in the process. Everything below assumes you reached a person.

Most teams have no standard prep

You might assume companies train people to run interviews. Many don’t.

In practice, your interviewers may be much less prepared than it seems. Their prep might look like this: “Here’s a rubric from three years ago, figure it out.” Or: “Let’s grab a conference room between meetings and decide what to ask.”

The questions are often whatever the interviewer personally studied when they were job hunting. These days, they may be generated with an LLM the morning of.

Then the panel negotiates. One person wants to quiz candidates on data structures and algorithms for a role in which they design websites. Another insists system design is essential for a junior level position. People default to what was done to them and assume it’s normal because it was normal to them.

What’s normal to the spider is chaos to the fly.

“Scoring” that isn’t really scoring

After an interview, some processes I was part of had one simple scale to score candidates: yes, no, strong yes, strong no.

The result is predictable. Like the candidate? Strong yes. They rubbed you the wrong way but answered everything correctly? Somehow a soft yes at best.

Structured scoring with defined criteria measurably reduces this. The research backs it, and the rare times I saw it used well, it changed my own assessments. Yet many teams I worked on never used this approach.

Prestige bias and politics

Even with a strong scoring system, bias and office politics can change the outcome.

For instance, I once interviewed someone I was strongly against hiring. It was clear they didn’t know what they were doing, and they’d be running critical infrastructure. I gave a strong no with objective reasons, scoring notes, specific examples from the technical round.

Leadership pulled me into a meeting right after and asked why. I walked them through my notes.

What I didn’t know: Several of them already knew the candidate personally. They liked them. They wanted them hired. I said the decision was theirs, my assessment hadn’t changed, and wished them luck.

I’ve also watched a strong resume short-circuit an entire loop. The team saw a top-tier company name, skipped the standard technical rounds, lobbed a few softballs, and basically welcomed the candidate in.

But once this engineer got started, it turned out to be a poor fit. And it wasn’t the candidate’s fault. They were set up for failure, because nobody checked whether this person could do this job at this company.

In both cases, it didn’t work out.

What you can actually control

You could read all this and decide the system is broken or rigged.

The broken part is fair. The rigged part isn’t. People who are genuinely good at interviewing pass more often. It’s messy, but it’s not a lottery.

You can’t fight bias, politics, or a sloppy process. That’s like being mad at the weather. You can only play the two cards you’re dealt: your technical ability and your behavioral presence.

Most candidates obsess over the technical side and forget the behavioral rounds exist. But product managers, designers, and cross-functional leads—people with zero technical background—will judge you entirely on whether you can tell a clear story and seem like someone worth working with. If you’re unlikeable in the room, you’ve roughly halved your odds at every stage.

So here’s the unglamorous advice that actually works: put yourself on camera.

Talk through a project you led, a mistake you made, a hard problem you solved. Record it. Watch it back. Cringe. Do it again.

Think out loud, under pressure, with another human watching.

If you keep failing interviews, the fix isn’t always more technical prep. It’s getting better at being in a room with other people who are potentially more nervous, less prepared, and more biased than you ever imagined.

The process is broken. You can still win.

—Brian

A new initiative from the U.S. National Science Foundation plans to distribute $1.5 billion of funding over 10 years to independent research organizations, which it calls “X-Labs.” The program is meant to support work being done outside of academic institutions, starting with two areas: scientific instruments for sensing and imaging, and interconnects and integrated photonics for quantum systems.

Read more here.

We’ve said it before, and we’ll say it again: AI is changing the engineering profession. So how can you stay in demand as the field’s tools evolve? A senior engineering manager at Walmart Global Tech offers seven quick tips.

Read more here.

For even more expert tips, check out the new career advice collection from The Institute. These articles feature guidance written by working engineers, meant to help those in all stages of their careers stay at the forefront of their profession. Discover tips for technical presentations, dive into a specific career path like cybersecurity consulting, and more.

Read more here.