Issue 5 / Failure

August 01, 2018
A salvaged 747 in the desert.

“The 747 Project,” an interactive art project at Burning Man made of a salvaged Boeing 747, created by a nonprofit called Big Imagination.

Life Aboard the Rocket Ship: An Interview with an Anonymous Engineer

Silicon Valley has a strange relationship with failure and success. It’s home to some of the most successful companies in the world yet it respects, even reveres, failure. It prides itself on an entrepreneurial culture where the vast majority of startups fail, yet it is increasingly dominated by huge corporations that are too big to fail. And what looks like success to Silicon Valley sometimes looks like failure to everybody else.

To help us think this through, we sat down with a successful engineer. They told us the story of their life, from learning to code to founding a startup to joining a big tech company. They talked about their failures and successes along their way—and how, as they grew disillusioned with the industry, their successes started to feel more like failures.

This interview was conducted in spring 2018.

How did you start writing code?

I went to a perfectly fine public school in Texas. We had a computer science course, but the teacher didn’t know anything about programming. We had a textbook, though, and if you were sufficiently self-motivated, you could work through the exercises and then take the AP exam. Fortunately there were a few of us who really liked to code. So we got together and taught the course ourselves. We all passed the exam, and it helped me get into an elite institution for college.

What did arriving at that elite institution look like?

It was weird. There was a big cultural gap.

I resented my family for not having prepared me for the experience. There were all these kids who went to private schools who were years ahead of me in math. I was angry that I didn’t get those opportunities, and that came out as resentment towards my parents for awhile.

I remember going to my college girlfriend’s parents’ house and feeling intimidated by the food they ate and the way they talked. But I also knew I wanted to emulate it. Throughout college, I put a lot of pressure on myself to make sure that nobody could say I wasn’t supposed to be there. That was the particular form that impostor syndrome took for me: it wasn’t about the fear of not being a good programmer, but the fear of not belonging.

The path that college put me on created some distance between me and my parents, though. They have never been anything but proud of me, but they stopped understanding me sometime in college. And in the years since, the gaps between our worlds continue to grow wider.

To this day, when I call my mom on the weekend, she still asks me if I’m off that day, because she doesn’t know that people only work Monday through Friday in offices. That’s just not context that she has—and I didn’t have it either before I left for college. My stepdad says if I ever wanted to move back home I could find work because there’s a local guy who fixes computers.

Apart from the culture shock, what was college like?

The school where I went wasn’t friendly to failure. It was a meat grinder. When you walked through campus around finals week you saw kids sitting on the ground crying.

Still, it’s the kind of place where you make connections that end up mattering a lot later. You meet people who will go on to start companies in Silicon Valley, or who will become higher-ups at bigger firms. It’s also the kind of place where recruiters track you from freshman year, have dinner with you at department-sponsored events—that sort of thing.

That industry interest made me feel special. And when I became a teaching assistant for a computer science course, that interest intensified. I had to do a work-study as part of my financial aid, and it was either serving food in the cafeteria or teaching computer science. But teaching paid well—$15 an hour, which blew my mind, since my high school job had paid $4 an hour plus tips.

Plus the job was super fun, and the industry attention felt great. Big firms would come in and lavish the teaching assistants with gifts—embarrassing gifts. Sometimes we would go buy cheap tricycles and ride them down a steep hill and play a sort of human Mario Kart. Microsoft found out about it and bought us the highest-quality tricycles available. I think it’s actually gotten worse since I’ve left. These days companies send stretch Hummer limos to pick kids up for fancy dinners.

So you must’ve been pretty well set up for a job in tech by the time you graduated.

After I graduated I had a few options of places to work. I chose an older, more established company. That ended up being one of the worst decisions I’ve ever made in my life. I spent two years there while my friends were jumping on rocket ships. Then I decided to start a startup with somebody I knew from school.

My cofounder and I had been talking for several months about doing our own thing. In college, everybody was expected to go start a startup—that expectation had been drilled pretty deep into us. We lived in a big house with a bunch of friends, and we turned an extra room into an office. We had a really nice routine. We worked there every day, and tried to prototype things very quickly.

This was around 2009, when mobile apps were taking off. It was the era of “SoLoMo”—“social local mobile.” The iPhone had changed everything: there had been programming platforms for phones before, but they were garbage. Everyone wanted to build something for the App Store. The tech giants were way behind the curve, however—the Facebook mobile app was horrible, for example. It seemed possible that the bigger players just weren’t going to make it on mobile, and that created opportunities for startups like us.

So we tried to build a bunch of different apps. None of them found a huge amount of success. But the last one we tried, a social app for making plans with friends, did well enough to get some attention from TechCrunch and other places in the tech press.

It was fun, but scary. Because we always knew that if one of the tech giants got their shit together, they could eat our lunch if they wanted to. I remember waking up one day and seeing a new product announcement from Facebook that I was convinced would put us out of business. I ran to my cofounder’s room, freaking out. It turned out we were fine. Still, I had that feeling of “Holy shit, we’re fucked” a lot. We lived in constant fear of getting scooped.

But you didn’t.

Not exactly. But after a year, we were still living off our savings. It was clear that we either needed to get funding, get a job, or get acquired.

Around this time, we went to an up-and-coming company to ask them whether they would give us access to their private API. They said, essentially, “No way. You’re a tiny company that doesn’t matter.” But they also said they’d be interested in acquiring us. We were very surprised to hear that. We didn’t even really know what it meant. But we decided to pursue it.

What they were offering is what Silicon Valley calls a “talent acquisition.” They wanted to buy the company, but they weren’t going to take our technology. They weren’t going to take our code. They just wanted to get the two of us working for them. They put me and my cofounder through an interview process, and it was the same standard interview process I’d already been through before. But at the end of it, they made an offer with a slightly different format. It was clear that we were talking to people who worked in their mergers and acquisitions (M&A) department rather than in recruiting. Those people have the authority to write bigger checks, and they’re supposed to be thinking about what the company is going to need a little further down the line.

We were very green, so we didn’t know what a good offer looked like. But we did know that we should go get at least one competing offer, to see what our options were. So, using our network of contacts from college, we reached out to somebody who had the ear of a few VPs at a bigger company that made similar acquisition offers. We said, “Hey, we have this offer from one of your competitors. They’re moving fast. Can we start a conversation with you?”

Recruiters famously don’t make the process very fast for people. You can do an interview and not hear back for a month, for instance. It’s frustrating for most people coming in the front door. But it’s very different if you have an offer from a competitor. An offer from a competitor is always the best way to get their attention, even if they have ignored you in the past. And doubly so through the M&A route.

How old were you at the time?

Twenty-three. This was all very new to me. Even with some experience in the industry, I had no idea how corporations worked. No one in my family worked for a corporation.

So we went into the bigger company and did a presentation for several directors. They found it interesting, because we were thinking about the same problems that they needed to be thinking about. Because the big companies missed the boat on mobile, they were willing to write checks to make up for that gap. That’s probably the biggest phase of talent acquisition that I’ve seen in my career. Although maybe today acquisitions around AI could rival it now.

They liked our product enough to make an offer. But their offer was many multiples larger than the first one we’d received. They were a larger company, with more money to spend. They were also more frightened of smaller competitors outmaneuvering them, so they were willing to spend more.

They wanted to acquire all of the assets of our company for a particular price and then wind the company down. The price they proposed was pretty high. And being a young kid with lots of student loan debt, I was blown away by the seriousness of that number. That’s the main thing I remember.

How did they decide how much the assets were worth?

It was a charade. Our assets weren’t actually worth anything. We had a rounding error number of users. Our service was not wildly popular by any means. So they valued the assets pretty arbitrarily—this chunk of code is worth this many millions of dollars.

You said the smaller company wanted to acquire you as a way to hire you and your cofounder, but wasn’t interested in your technology or your code. What about the bigger company?

They were planning to throw away every line of code. There was nothing that they were actually acquiring besides us.

Then why buy your code if they’re just going to throw it away? Wouldn’t it have been easier to just hire you, instead of hiring you and buying your code?

Sometimes the big tech companies acquire startups to acquire their technology, and sometimes they do it just to prevent those startups from becoming competitors. In our case, it was the latter.

Even if a big company is not directly threatened by a startup as a competitor currently, the thinking is that if they need to buy them later, they’re going to pay a lot more for it. So they might as well buy the startup as early as possible to nip it in the bud. Our startup was pre-funding, so they could get away with paying us much less. We didn’t have any investors they had to satisfy.

But acquiring our assets was also a way to justify paying us a lot. If they’re only going to pay you 2x the normal salary, then that can take the form of a very nice job offer. But if they’re going to pay you more like 8x or 10x, it breaks the whole idea of salary bands, which is how big companies organize compensation by experience level. So buying your assets is the backdoor—it’s a way to get away with paying certain people much more.

As far as what they’re buying—yes, they’re avoiding paying more for a potential competitor later. But the inherent value in a talent acquisition comes from acknowledging that most projects in software fail. Finding a team that can actually ship something that gets out the door is rare. Even at big companies, most projects will not see the light of day. So to find a group of people that have managed to build something—even if it’s small, even if it’s humble—means they’re probably a team that works well together. So they’re worth a premium. That’s the theory behind it, at least.

Also, they could make us sign a contract that locked us in for a long time. The deal to acquire our startup was a lump of cash and a job offer. We had to take both together. About half of the payment came up front, in the form of the cash. And the rest would come to us through salary and stock-based compensation on a vesting schedule over the course of four years.

Sure, I could’ve showed up on day one and quit. And they would’ve been angry at me, but I still would’ve been able to pay off my student loans. However, I would’ve been leaving a lot of money on the table.

Were you excited? Paying off your student loans must’ve felt pretty good.

I was very excited and very terrified. I didn’t want to screw it up. The deal was complicated. There were hundreds of pages of legal documents that I felt very overwhelmed by. We had to pay a lawyer fifty thousand dollars to make sure everything was airtight. My parents didn’t understand it, and to this day don’t understand it. My friends in tech were happy—some of them had been through this experience before.

But it changed my life. I went from being basically broke—my next rent payment would have emptied out my savings account, not to mention my student loans—to not having to worry about money anymore. So that was great.

How did it feel to go from running a startup to working for a big company?

It was intimidating, but there were some really positive aspects. As an engineer, I learned a lot. I felt like I was finally learning how to actually write software. I would go home and read the company’s internal wikis for hours. I was so excited about working there that I read documentation every weekend for a year, actually.

But coming in as a talent acquisition, you’re also expected to be a thought leader. You’re expected to inject the company with new ideas. It’s an informal role—it’s not reflected in your title. But that’s why management has paid a premium for you.

So my cofounder and I started to materialize what we thought this company needed, and assembled a team of people to work on it. The fact that we could come in and substantially change product direction—that we could create and staff and launch a project—was due to the fact that the company had paid extra for us. Leadership assumed we knew something.

Failing Up, Down, and Sideways

You said that most projects in software fail. Why is that?

Because you never know what’s going to work. Market timing is everything: something that makes perfect sense two years from now, or made sense two years ago, might fail today for no good reason. Everything changes so fast: the technology stack, consumer demand, even the fundamental capabilities of these devices.

I mean, everything I’ve ever worked on has failed. I’ve worked on some ambitious projects at several of these big companies, and none of them have succeeded. But I’ve still been rewarded and promoted. And I think that’s a good thing about Silicon Valley. Failure isn’t looked down upon, which is a positive aspect of tech culture.

When you fail inside a big company, does it still feel like failure?

The average time spent on a team is well under two years at most of these big companies. So when a company wants to change direction and abandon a product, people usually don’t take it that hard because they weren’t planning on being there for very long anyway.

On some teams, however, that’s not the case. I’m currently on a team that has been working on something for several years. And it’s failing. We have been launching small representative parts of our product but users aren’t using them.

This is partly a problem with what constitutes success within a big company: if you launch a product with a million users, it’ll get killed because a million users is nothing. That’s one of the reasons that big companies have trouble innovating, because achieving a 1 percent gain in users of your main product will win every time over launching a new product with a much smaller user base. I mean, a million users would be a rocketship success for an early startup. For a big company, it’s a drop in the bucket. That’s why big companies tend to get stagnant, because they’ll always prioritize growing the main product over funding experimental ventures.

Anyway, on my team, failure has really depleted our energy. We’re demoralized because we’ve been grinding for a long time on something that just isn’t taking off. It used to be one of the best teams I’ve ever been on—but within a period of six months, we’ve become very unproductive. We have no direction.

That sounds like burnout.

People get burned out not because they’re working too hard but because they’re not feeling rewarded by the work they’re doing. They get burned out because they believe their work has no impact. On my team, since we know it’s only a matter of time before leadership kills our product, people are burning out left and right.

At work, there are certain things you have to do. But what the company is really paying you for is to come up with new things to do. They’re paying for your creativity. When I’m burned out, I’m still doing the things I have to do—I’m filing the TPS reports—but I’m not coming up with new things to do. Burnout is when the creative part of your work is dead. There’s a muscle I go to flex and it’s just not there.

What about in the startup world? Failure must look different inside a startup.

If you launch a startup that goes out of business, no one thinks you wasted your time. People still revere a founder whose company has failed, even to a fault. Plus, because there’s always more money being pumped into tech, it’s a soft landing for almost anyone whose startup fails.

That changes depending on what you’re working on and the time period, of course. For instance, I don’t think if your social app failed today you would have a nice acquisition offer waiting for you, unless you knew the potential acquirers on a personal level. And that ends up being the major way that the opportunities for “failing upwards” are not distributed equally.

Who is allowed to fail, and who gets to fail upwards? Your startup wasn’t failing exactly, but it sounds like the social network that you acquired in college was the determining factor in your ability to land acquisition offers.

Definitely. In our case, that was the main thing. We didn’t have an impressive piece of technology or an impressive user base. But we did have social capital.

There are other ways to acquire social capital. If your startup gets press attention, that raises its acquisition price. If you have a really stellar team, that’s another way to fail upwards. You could assemble a dozen excellent engineers to work on a very hard problem and then fail at that problem. But you found a dozen engineers that can work together without killing each other and maybe even manage to ship something. That’s worth a lot.

On the one hand, Silicon Valley seems to revere entrepreneurialism. On the other hand, the industry is increasingly dominated by a handful of big companies—companies that, as you’ve explained, frequently acquire startups and burn down all their assets to ensure they don’t become competitors.

How do you make sense of that contradiction between the cult of the founder and the increasingly monopolistic structure of tech?

The funding model for startups is venture capital. And venture capital is a hits-driven business: you expect the vast majority of your investments to fail, so the ones that succeed have to succeed on a massive scale. Venture capital is risky, and it requires a lot of money.

Until relatively recently, tech companies didn’t have enough money to compete with venture capitalists. But now they do. Today, you have four or five tech giants with cash piles big enough to really push people around. And this has only happened in the last decade or so—it wasn’t like that in the early 2000s after the first dot-com crash.

But the incentives of a VC firm are different from those of a tech giant, right? The former is giving you capital to help you grow into a bigger company, whereas the latter is buying you to make sure you don’t grow into a bigger company.

Right. But again, the tech giant is also buying you because you’re a founder who has had some amount of success. The reverence for the founder might sound silly, but it’s based on something real, which is that it’s really hard to measure why a company is successful.

There are too many factors at play: market timing, staffing decisions, choices about the technology stack. It’s impossible to know why a particular startup succeeds. So you find something to control for, and that’s the people. That’s why big tech companies like to bet on founders. This is something I’ve noticed in my career: there are people who are just very effective, and the things they touch seem to work. And most people aren’t like that.

How have your views of the acquisition experience, and of the tech industry more broadly, changed over time?

I felt I was succeeding while the acquisition was happening. I had worked hard and I was being rewarded for it. The system was working.

But today I look back and think of it as a failure. Why didn’t I work on something more challenging? Why didn’t I take a bigger swing? At the time, I thought I was working on something groundbreaking. Now, with some hindsight and maturity, I think I just got lucky doing something small.

I definitely feel like this was a silly chapter in my life. I’m very glad I got to pay off my student loans. But I don’t feel good about the work. In fact I feel pretty embarrassed about it.

It’s not that I consider myself a failure for not having built a successful business. The failure I feel is more personal: it’s that I spent my time on building a small social app instead of something that would have been more meaningful.

What would’ve been more meaningful?

I could have worked on something that actually had a positive impact on society. Or I could’ve set out to solve a problem that had more interesting technology behind it.

Making the World A Better Place

What do you think changed your mind? Because you didn’t feel that way at first.

I think it was seeing more of the industry, spending time at these big companies, and observing certain things.

One of those had to do with bias. Silicon Valley has been under attack for the past several years for having a lot of bias in its compensation practices around gender and race. As a result, the big companies have tried to standardize the way they pay people. They’ll break compensation into salary bands that are supposed to match an employee’s level. So if you’re at a certain level, your pay falls somewhere within the corresponding salary band. Which means that if a man and a woman are in the same band, they’re going to be paid roughly the same.

That’s supposed to help correct gendered pay discrepancies. But it doesn’t really work, because there are all sorts of escape hatches built in. Salary bands only cover your salary. There’s lots of other ways that people get paid.

As we discussed, talent acquisition is one of them. Talent acquisition gives companies a way to pay a premium to people who have more social capital. But that’s not the only way that people are rewarded unequally. There’s also the sign-on bonus. The sign-on bonus in Silicon Valley today can easily be a hundred thousand dollars. Even for somebody coming off their first job, or maybe even right out of school, it can be upwards of fifty thousand dollars. And the recruiters have a lot of leeway in setting that number. Then there’s your annual bonus, which is a percentage of your salary at most companies. Finally, there’s your stock-based compensation.

When you take an offer at a company, you’re given either stock options or grants of shares in the company. Those options or grants vest over a four-year schedule. And there’s really no restriction on how high that can go. So for a lot of people, a majority of total compensation comes from stock. Salary typically tops out at around $200,000 or $250,000 at a big company. But it wouldn’t be surprising to be given another $100,000 in stock grants. If you’re joining a company early on, that stock, by the time you’re done vesting it, could be more like a million dollars.

What about ageism in the industry? Silicon Valley tends to be very young. Does that worry you as you get older?

I don’t know where all the old programmers go. They must go somewhere. It is a little worrisome. I’m in my thirties, and I feel like I have less energy. I’m an iOS developer and I haven’t learned Swift. Five years ago, definitely ten years ago, the day Swift was announced I would have become an expert on it. And I just don’t have that energy now.

So it sounds like issues around equity and bias played a determining role in changing how you think about the industry.

Well, come to think of it, those are issues I’ve only come to understand in the past few years.

I think what really shifted my thinking about my success was observing that most of the startups that are getting crazy amounts of venture capital aren’t solving interesting problems. There is a lot of money going into squeezing more ad dollars out of users, and getting more attention and eyeballs. Yo is a good example. My app wasn’t much dumber than Yo. And they got millions of dollars of funding.

I also felt like people were becoming founders and investors not because they wanted to solve problems that would help humanity but because they wanted to be in the Silicon Valley scene. They wanted the cultural cachet. They wanted to go to the parties.

That was disillusioning to me, because I had really bought into the ideology that building a business was the best way to make the world a better place. That was something that was drilled into my head at my elite college. I had completely bought into it.

But over the years, as I saw what products were being funded and built, I felt disappointed. It changed how I thought about my own success. I wasn’t actually solving problems—I was just riding a wave of ridiculous overinvestment in social apps.

So your own success started to feel more like a failure—not a personal failure, perhaps, but the failure of the industry more broadly.

Recently, you’ve actually had a number of high-level people in Silicon Valley express some degree of disillusionment as well. Early Facebook investor Sean Parker, among others, seems to regret his role in building a platform he now considers psychologically damaging. How do you see this disillusionment playing out more broadly in the industry?

Get wealthy and solve the world’s problems—this was the message that I absorbed early on. You couldn’t do one without the other, the argument went, so don’t feel bad about becoming rich. The profit motive is the only way that we can possibly solve problems at scale.

I truly believed that. And it remains a widespread belief in the industry, and in the engineering departments of elite institutions. But to move forward, I think we are going to have to challenge that belief very directly. That’s why I’m skeptical of some of these newfound regrets expressed by Sean Parker and others. I don’t think they’re actually attacking the core notion that the profit motive is the best way to make the world a better place. They still believe they can centralize large amounts of capital in these massive corporations and pay themselves well and solve the world’s problems. I think there are some inherent tradeoffs that they’re not yet acknowledging.

Sean Parker is also a billionaire. Do you see rank-and-file tech workers expressing doubts about what they’re building as well?

When you’re an engineer, you’re constantly being told to do things that are clearly not good for the user. If you’re building any kind of app or platform that makes its money from advertising, you are trying to maximize “time spent”—how long a user spends with your product. Time spent is considered a proxy for value delivered to the user, because if your product wasn’t useful, the user wouldn’t be using it.

Here’s how it typically works. An order comes down from on high: the board says to increase revenue. What’s the best way the management team knows to increase revenue? To increase time spent. So they issue the order, which gets percolated down the tree, and now everyone is working on increasing time spent. This means making the product more addictive, more absorbing, more obtrusive. And it works: the user starts spending more time with the product.

But every worker knows this is bad. Every engineer and designer knows this is awful. They’re not happy making these features. But they can’t argue with the data. The engineer and the designer who care about the user don’t want to put these features out in the world. But the data says those features are increasing time spent—which means they’re good. Because more time spent means selling more advertising, which means making more money.

And so long as you’re working for an advertising company, what other metric besides time spent could there be?

So long as you’re working for a company, what other metric besides profit could there be? That’s a similar question. You can make small surface-level improvements here and there. But you’re not going to tackle the core problem until you tackle the profit motive.

The directives to increase metrics like time spent come from above, but the actual work is being done by tech workers on the ground. And they’re doing this work because their performance is measured by whether or not they moved that metric and whether or not they implemented those features—even if they know they’re bad for users.

But there’s no way they can push back on it. They can talk about it—in their company Slack, in their public forums, at their all-hands meetings. They can express a lot of malaise about it. But they can’t argue against the experiment succeeding, because you can’t argue against increased profits.

You could imagine different structures of the company that might not have this problem. You could imagine a world where these companies empower rank-and-file workers to make certain decisions themselves, and give users a voice in those decisions. Workers and users could together decide what metrics to optimize for, and what kind of technology they want to build.

This piece appears in Logic's issue 5, "Failure". To order the issue, head on over to our store. To receive future issues, subscribe.