Fear rules everything around me

Content warning: brief mention of suicide + generally heavy mental-health things.

One of the first engineers we worked with - let's call him Bob - booked a call with me, then canceled it. He scheduled another, then canceled that and rebooked again. I can’t tell you how sorry I am about this, he wrote in an email.

Bob canceled again, and at this point I followed up with him to ask what was going on.

I've been working as a freelancer for personal reasons, and now I’m trying to look for an opportunity for a full-time position. So I'm a little nervous and feel I have to be perfect for everything.

I did my best to reassure him, and rescheduled. Bob canceled the next booking, and the next. After the fifth cancellation, I told him - for his own good as much as ours, since he was clearly going to keep kicking the can down the road - that he should make sure to come to this one, and that we wouldn't rebook him further.

Bob didn’t show, and that was it.

Not a week goes by when my job doesn’t turn into that of an unofficial therapist or coach. Sometimes it’s run of the mill anxiety: is your interview hard? (Yes.) Is there anything I can do to prepare? (Yes.) And sometimes it’s people on Reddit whose dad killed himself on their birthday twenty years ago and who are scared that that's undermining their ability to succeed in interviews now.

No, I'm not making that second one up.

Bob is not alone. If you’re like Bob, you’re not alone either. I’ve got no less than five candidates right now putting off our interview out of anxiety. Well, either that, or having a high-pressure interview booked is a severe risk factor for sudden deaths in the family.

I wish I could tell everyone who’s afraid that they’ll be fine. Our interview is easy! You'll definitely pass! It’ll all be sunshine and rainbows, because you’re great! You got this!

But that wouldn’t be true.

Most people who take our interview - about 70% - fail it. Here's a breakdown of current results pulled from one of our internal dashboards, which is omitting a string of (largely awful) early interviews that weren't tracked in this dataset. Most of the missing interviews would be in the dark red section of this chart.

Around a third receive feedback that looks like this, with not a green pixel or check-mark to be seen.

On one level, I hate writing feedback like this, especially for candidates that already seemed stressed. It feels cruel. “Here,” we say, “is precisely why we are closing the door in your face. Good luck out there in the cold!” Someone came to us for help, and we're turning them away.

But I don’t think dishonesty helps candidates. If someone were to bomb our interview, and if we were to pretend that they didn’t, we wouldn’t be doing them a service. We’d be misleading them about where their problems are and where the solutions might be. We should give them that truth as compassionately as we can, but we promised them facts, not platitudes.

And for the same reason, I can’t tell an anxious person that they’ll be fine. 70% of the time, they’ll get a rejection. 30% of the time, they’ll get a detailed inventory of exactly how they suck.

No, here’s what I want to tell them: with high probability, you will fail our interview.

You should do it anyway.

This page explains what you already know

Let’s go back to Bob, our six-cancellation guy from the start of this post. 

Psychologists have a name for Bob's pattern of behavior: avoidance. The interview feels scary and stressful, so Bob avoided it out of the perception that he'd feel better. I don’t want to think about the cucumber that’s been slowly going bad in my fridge for weeks, so I just pretend it isn’t there, because thinking about it feels like it would be more stressful than avoiding it.

I think people who are less acquainted with - and less sympathetic to - the realities of mental health than I am might say “man, what a wimp, Bob should just tough it out!”. And there is a sense in which that is true. He should, in fact, do that, all else equal. But what this response misses is that Bob already knows that. The problem is that he isn’t able to do that yet. He tried to tough it out, but he did not succeed.

Mental health advice that includes the word “just” is missing the point entirely.

If you're like Bob, an interview isn't just an interview in your mind. It's a referendum on your future, on whether or not you're going to have the career and life that you want to have. It's a judgement of your worth as a person, of whether you have any value as a professional. You're not afraid of losing this job. You're afraid of being a failure, in a much grander and more general sense.

From the perspective of a recruiter, it's easy to see that this feeling is common. But because admitting that you're not sure you're good enough is a terrible idea in a job hunt (seriously, don't!), and in a professional environment in general, you might not be aware that if you feel this way you share it with at least a third of people on the job market, and probably much more than that.

Bob's problem is not one of knowledge. Bob's problem rests in his feelings and his mental processes and not in his logical judgements. To do the thing, Bob needs more than the knowledge that he should do it. He needs to feel like he will feel better doing it than he would avoiding it.

Fortunately, avoiding things feels pretty bad! Or at least it does to me. I hate the feeling I get when, at the end of a day, a thing I've been diligently not thinking about slides back into my consciousness right before bedtime (or when a deadline forces me out of avoidant thinking). Judging by the distressed tenor of Bob's apologies for his cancellations, I'm pretty sure Bob felt awful, too.

But Bob avoided himself out of a job anyway. The fact that Bob continues to avoid things means that doing them must feel worse.

But why? How is it that Bob can feel worse about taking a chance to get a job than he feels about guaranteeing that he won't?

Fear is unbounded, consequences are not

The fastest way to drain the tension of a horror movie is to show the monster on screen.

A visible monster’s scariness is constrained by reality. But an unseen one exists only in your mind, and is constrained only by your mental processes. The space of possible mind-states is larger than the space of mind-states that correspond to reality. That's why you can draw Penrose steps and have the spatial sense of infinite stairs, even though they can’t exist in reality. A person on psychedelics can see impossible colors that cannot be produced by any physical stimulus but that are perfectly permitted by the neurons in the brain, because what they "see" is their own brain circuity, not the light entering their eye.

Similarly, while the actual consequences of failing an interview are constrained by the laws of business and economics, the fear of failing an interview has no such limitations.

Bob was not thinking about the worst he could possibly do on our interview. Instead, Bob was thinking about the worst possible future his brain could imagine, with a bad result on our interview as part of it.

I do the same when I feel anxious about how my company is doing. I don’t imagine simply not succeeding as a founder. Instead, I imagine the worst outcome for my life, with the failure of my company as part of it. I imagine a world in which I'm too weird to be successful, in which trying to do business honestly makes me a naive idiot and not a virtuous freethinker, in which my difficulties forming and preserving relationships doom my business ambitions, in which I slide back into a depression from which I never recover, all because this one failure would "prove" all the others are inevitable. This is, of course, not true, or at least it is only one of many possible paths. But it's what my brain wants to show me.

Bob and I aren’t just imagining the worst case. We’re imagining worse than the worst case. You might be, too. And that means that it can help to stop and just ask what the worst case actually is.

In Bob’s case, the worst case outcome of our interview is that he would not get a job through us. But that isn’t really that bad an outcome. He spends a little time and energy, and it would feel bad, but it isn’t a disaster. It doesn't put Bob in a meaningfully worse place than he was before. And importantly, it’s no worse an outcome than just not taking the interview, which guarantees he won’t get a chance at a job and guarantees he'll feel terrible because he knows for sure that the "failure" here was his and only his.

And Bob is trapped in a loop. When he avoids the interview, he avoids this lesson. He avoids looking at the monster. In so doing, he avoids putting any constraint on how terrifying it can be. The fear he feels can feed on itself, unconstrained by reality.

It is in this sense, and only in this sense, that the "just get over it" advice is useful. It's not that you have to not be uncomfortable. It's that you have to develop the skill of looking that fear in the eye and forcing it to interact with a reality that is less terrifying than the anxious mind will make it. You don't defeat the discomfort: you reduce it to a form you can process and manage.

I can't make you less afraid right now. But I can tell you that if you turn and look right at the monster, it will not be as scary as you fear. The more you look at it, the less bad you'll feel.

Don't fail our interview because you should fail it. Fail our interview because you will feel better afterward.

Even low-probability bets win sometimes

On June 25 - less than two months before this writing and on the eve of what ended up being probably the most consequential Presidential debate in history - PredicitIt users were pricing Kamala Harris’ chance of winning the 2024 Presidential election at 3 cents. Given the fee structures involved, users were setting Harris' implied probability of victory very close to zero.

Well, we all know how that turned out. As I write these words, Harris is priced at 56 cents. Other markets are a tad lower (as of this writing, Manifold prices her at 54 cents and Polymarket at 48), but they all agree that she has approximately coin-flip odds of victory. If you don't trust prediction markets, first-principles models show more or less the same thing, and pundits on both sides don't seem to think those models are wildly off-base.

On the other side of the aisle, eight years ago it was pretty hard to construct a narrative where Donald Trump would complete a hostile takeover of the Republican Party. In fact, it was extremely easy to construct narratives about why that wouldn’t happen. We all know how that turned out, too.

I raise these examples not to make a political point, but to make a practical one about the difficulty of predicting low-probability outcomes.

It wasn't that, taken literally, either of these outcomes was impossible. I think if you told the story of what actually happened to a reasonably empirically-minded person before they actually did, they would have conceded that the sequence of events was not forbidden by any law of physics. But with the information available at the time, it was much easier to construct narratives where these outcomes did not happen than it was to construct outcomes where they did.

These were narratives written by very smart people, experts in their domain, based on historical evidence and years of focused study. They were not bad narratives. They were sound bets that, at most, perhaps underestimated a 5% chance as a 2% chance or something along those lines.

But a sound bet is not a certain bet. And conversely, a low probability bet is not a certain loss.

I can tell you half a dozen reasons my company won't work, and paint you a hundred futures where it fails. It takes a lot more work to construct the narrative for how it would work. And so I systematically underestimate the chance of events for which I have no narrative - like, say, waking up in the morning to find that some random person you don't know has put your blog on the front page of Hacker News, resulting in half a dozen new clients and more than a hundred very-high-quality candidate signups. One of the reasons my previous post stressed the logic behind failed decisions was to try to make this point.

The same goes for interviewing.

We interview engineers all the time who (a) appear to be rather terrible engineers and (b) have a long history of being employed as engineers. (These engineers, by the way, are often not afraid at all - and it's not to their benefit!) One easy explanation is that if you simply take enough shots, you eventually get lucky. You get asked the exact questions you know. You get an interviewer who’s in a great mood. You happen to be a perfect fit for what that company is looking for. You talk to a company that is desperate.

All of this is just another way to say “you miss 100% of the shots you don’t take”. But that platitude can be unpacked into two more concrete, more actionable statements.

First: if the value of success is high and the cost of failure is low, the expected value can be positive even if the chance of success is small. This is the principle on which startup culture runs: an MVP is cheap to build and its potential upside is enormous.

And second, human beings are bad at estimating tail probabilities, especially in environments where our information is limited, our sample sizes are small, and our narrative imagination is limited, so the chance of success in a low-probability bet is often higher than your intuition would suggest. And this is why you build an MVP, rather than just talk to users: sometimes outcomes occur that are outside of your narrative imagination.

So while this is not exactly a reason to fail our interview, it’s a reason to do it even if you expect to fail. The benefits of success are high, the cost of failure is low, and you are probably underestimating the probability of success. Bob might or might not have passed, but he was probably more likely to pass than he thought he was.

The people who did fail it are glad they did

So far, this whole post has been about why you should fail an interview, not about why you should fail our interview. That’s intentional, both because most of this has nothing to do with us as a company and because making your blog posts thinly-veiled pitches for your company's offerings is for chumps that have nothing important to say.

But we do have our own data to bring to the table, and I want to conclude with that.

I went back and looked at all the recent replies I've gotten to rejection feedback. I want to emphasize that this is not a selection of cherry-picked quotes. This is every recent response to a rejection that was more than a quick “thanks”. I've edited it only for brevity and to remove some identifying details about the candidates and our interview content, not for sentiment.

The experience and feedback have indeed been helpful. You've given me a lot to work on and I appreciate the opportunity despite the outcome.

Wow, I'm grateful for such a comprehensive and thoughtful response--it took some time to chew on.

Thanks a lot for the feedback. Can't say I'm surprised about the feedback about my tendency to wander off on bunny trails. It's something I'm aware of, and I agree, it was really bad on this interview…Anyway, thanks for the interview, I'll see if I can get better at this!

Thank you for this feedback, it’s very valuable! 

Woah! That was fast and the feedback is spot on…Please sign me up for your newsletter.

Appreciate the opportunity. The interview gave me a lot of perspective on things I need to work on. 

Thank you and [interviewer] very much for the feedback. It was quite the reality check. I'm glad I still retain some basic knowledge, but I didn't realize I forgot so much more of it.

Hi Rachel. This is great, thank you so much, really appreciate the feedback. 

Now, is it possible that these people are just being polite and are actually pissed off at us, or that some people were miserable and just didn't reply? Sure. We have had people who were unhappy with our feedback before, and I would be shocked if no one walked away feeling bad.

But it's also possible that the obvious conclusion is true.

We didn’t get these people a job. We didn’t even introduce these people to a job. We interviewed them and then told them why they weren’t good enough, and that was it. But they learned something. They walked away with more than they had before. They looked at the monster, and they saw that its teeth weren't that long, its claws not that sharp.

It wasn't a disaster.

And none of them seemed to regret it.