Hacker News new | past | comments | ask | show | jobs | submit login
What happened when my co-founder quit the night before our YC interview (brandonb.cc)
381 points by brandonb on Nov 14, 2013 | hide | past | web | favorite | 92 comments

Six days before the interview, my laptop crashed and we lost our entire demo.


I'm glad you mentioned how you lost your work because of lack of a backup. There are quite a few snarky comments here about it, mine won't be one of them. Instead I thank you for your candor.

There are some cultures where mistakes and failures aren't explicitly mentioned; they're swept under the rug (wouldn't want to lose face!). I hate that. Instead I think we can all learn a great deal by reading someone say (in not quite as vulgar words): "I fucked up. I made a very simple, stupid mistake". It's of course implied the takeaway is "don't be as stupid as I was".

Yeah, agreed on that one. When I read that line, my first thought was, "GAH! Why wasn't he storing this in source control somewhere remotely?!"

And then I stepped back for a second and remembered that just 4 months ago my laptop was stolen and I lost 6 months worth of photos... which of course I hadn't backed up anywhere.

We all make mistakes. We all forget things in a time crunch that don't seem so important until something really bad happens.

Indeed, I've always been insanely over protective of data (since I had a hard drive crash in the 90's take out an essay I was writing for school) but for some reason I never applied that to my phone.

Then my phone and wallet got stolen and it was a massive mess as it had access to just about all my critical stuff (dropbox, google apps, email accounts etc) and I had no remote wipe and no password on it (yeah I know).

Now I have both remote wipe, a strong password and very little sensitive data on my new phone, far too easy to lose the keys to the kingdom.

If you were to encrypt your phone, you'd not have to worry about it being stolen with sensitive data on it.

If you are using remote wipe without encryption, all the thief needs to do is drop it in a foil bag.

Only if the phone was off when it was stolen, or it was on, and you absolutely trust the security of the screen lock, which I'd be wary of.

And thank you for this comment.

You point out one of the most poisonous things about nerd culture. By pointing out when people are doing something that is in retrospect stupid, we train them to hide mistakes.

I've done it plenty myself, so I'm casting no stones here. But there's a giant gap between appearing smart and being smart. We rightly value smartness, which is great. But by playing Nelson to stupid mistakes and stupid questions, we encourage the appearance of smart over actual learning. To keep getting smarter, you have to be willing to look stupid sometimes.

I don't know when all this happened, or when this interview was, but while I'm certainly not always 'backed up', there really is very little excuse for loss nowadays.

Between Github, S3, Dropbox, etc., it just about takes effort to be in a position to lose data.

That said, I don't mean to fault the author as much as I want to encourage everybody to start using the tools available as part of their every day routine. If you're in a position now where catastrophe to a hard drive or a stolen laptop could affect your business in any way, you're doing it wrong, and it takes practice and regiment to correct, so start correcting it now, or you never will until it's too late.

Just to let you know there companies with cleanrooms that can get back the data of your HDD. You will have to pay a big bunch of money but in these cases it worths it. And they are really efficient. I saw a guy doing it for the cloud of his company, almost all HDDs are saved even one that was burned by a lightning strike.

I'm the OP. This happened two years ago, but I figured I'd post it now since YC interviews start tomorrow, and I already heard of one person in this batch who's going through the same thing. Good luck to everybody interviewing! If you have any questions, I'm game.

Six days before the interview, my laptop crashed and we lost our entire demo.

I am constantly shocked how intelligent technology people frequently fail to make backups.

Backups are so easy today - USB sticks, cloud backup, server backup, etc. Why do so many people ignore this basic step?

It's kind of like eating your vegetables and exercising. You know you ought to. But most people don't.

You'll never lose money betting on aggregate laziness and lack of foresight.

Just put all your project files in Dropbox!

Seriously, it's so easy. Every time you hit save, you'd have to try to lose the file. Pair this with git going to github or bitbucket and it would take several well placed nuclear explosions to destroy your work.

If my laptop instantaneously combusted at any point in time I would, at maximum, lose approximately 6 seconds of work (or the last time I hit "save"), and could set up a new laptop and be productive on it in about an hour (to install Dropbox and give it time to sync, Sublime, iTerm, Chrome, and custom keyboard hacks). Pet peeves, man.

I used to do this, but then when compiling the project the Dropbox client would shoot to over 70% CPU usage and the fan on my laptop would go into overdrive.

Switch to GitHub/BitBucket

This is due to churn on the compiled files right? The fix would be to compile these files someplace outside of dropbox. (Because the important artifact you want to back up is source code, not binaries). Depends on your language / build tool for how complicated this would be...

CMake encourages out of tree builds. I find them superior to in tree builds in most cases. But if it is just a random utility that I'm compiling from the Internet and do not intend to dev on, in tree works fine for me.

A suggestion off the top of my head would be writing a script that pauses Dropbox syncing before running a compiler. (From a quick Google search, there seem to be a couple of scripts already written that do this.)

Dropbox has been a life saver for me. The ability to restore previous files in particular has helped me in many situations.

Well-designed backup systems (say, Dropbox) don't actually require any overcoming of laziness.

You'd need to install Dropbox and then also save your project in your Dropbox folder.

Dropbox isn't a backup solution, and you're doing two steps when you could just do one - install backup software.

I can't imagine writing code without pushing to a remote git repo. Bitbucket is free!

This was my thought as well, pushing to a remote repo is that moment after which I stand up and walk away from my computer. I can stretch, make coffee or whatever. If I get back to my computer and it has exploded I know how far I am.

Heck, without git how do I revert changes? How do I stay organised?

Its just a property of humans I guess. If it worked the last 600 days, why would it fail tomorrow?

And thank goodness a lot of people have ignored that.

Back in Ye Olden Dayes, we trained everybody to back up. It was an important ritual: every once in a while you had to stop whatever you were actually trying to do, spend a bunch of time fucking around with odd, expensive media, and the store the media in complicated ways.

It was a giant pain. Most people didn't do it. But we programmers, wanting people to be as programmable as computers, told them they were doing it wrong, that they had to learn the rituals to pacify the machine gods, who would otherwise destroy their data.

I'm glad they ignored us, because without them we wouldn't have moved toward the correct solution, which is making the computers do the work to solve the problems created by computer use.

We still haven't reached the logical end, which is where every consumer computer is backed up by default on every change. But we'll get there eventually, thereby freeing up people to spend their time on what really matters. Which isn't wasting that intelligence on making backups.

Are you the dude who did the "we're going to make it so nobody's going to fuck with our customers" line at demo day? Cause that was awesome. :)

Indeed, that was my (new) co-founder!

Looked at your website, wish I could see some kind of demo or case study.

What was the original idea that got you into YC?

A mobile app for local events. To put it lightly, YC strongly encouraged us to change the idea to a better one: http://brandonb.cc/no-filter-the-meanest-thing-paul-graham-s...

This happened to me too.

My cofounder quit the day we got our email to interview and wouldn't budge no matter how much I tried to reason, cajole, or beg him to change his mind. At the time I played a weekly poker game hosted by a bunch of YC founders and I still remember walking in with two bottles of champagne in my hand that I bought before I got the breakup text and announcing "guys, I've good news and bad news and reason to drink either way".

I don't blame him at all. We had built up a ton of great chemistry working together for 6+ months but it was on a startup that flamed out with a firesale acquisition and he wanted more stability, especially with his wife grueling away at a residency to pay for their mortgage.

I still think it was a mistake though and not just on his part. He was my Woz and I would've gone to greater lengths to get him to stay if I truly had the Jobs zeal.

I bet you'll never forget the memory and ironic imagery of walking in with celebratory champagne for a startup that lost one of its founders to a room full of founders.

Also, would love an opportunity for a seat at one of those poker games.

Yeah, I miss that game. It ran for a couple of years before everyone moved on but a couple of us are thinking of putting it back together. Ping me at my HN username at gmail and I'll let you know.

This is one of the most inspirational posts I've read by any Entrepreneur and it's the character that you show by being able to follow through with the interview and then to write this post that tells me they made the right decision in picking you.

I happen to know of Sift Science and I feel its one of the stronger products in the valley today, but that said, even if I did not know your company at all I'd be rooting for you even more after reading this.

The good ones know exactly what to say ... The great ones say exactly what should be said, mostly without even knowing it.

Thanks for this post Brandon

Is this what HN has become? Almost the whole lot of comments concentrated on the lack of backup, and how he should have used dropbox or a USB Stick etc.

Come on folks, one or two comments are probably enough. Lets move on.

Nice story but seriously ?:

> Six days before the interview, my laptop crashed and we lost our entire demo.

No backups, svn/git/vcs, Dropbox, a USB drive?

Losing a partner is a difficult thing in business. Losing your code is just silly.

Yeah, not having a backup was totally stupid. Although, to be honest, it probably wouldn't even make my top 10 list of mistakes nowadays. Starting a startup tends to increase the number and magnitude of both your successes and failures.

That's what stood out to me also. This wasn't bad luck, it was stupidity.

Kudos on not giving up. I've had our co-founder/CTO quit on us t the day we were expecting term sheets. Pretty bad timing and I was freaking out a little bit.

Investors have seen it all. While I expected this setback to screw up our funding, it didn't make a difference in the end.

When this kind of shit happens, just carry on.

The thing is, if he was shaky, the earlier he disappears, the better. And if you can show you can go on even when losing key people, that's a good sign, because chances are high you will lose key people.

I fully agree. I actually realized only after that we're very happy he's gone. He wasn't a great fit at all. Even though he was an outstanding developer, we're much stronger without him.

Hmmm I was rejected this past April and then sent a reply back saying something like your loss.

Miraculously a week later I was invited out to Silicon Valley to demo our novel tech, though not to Y Combinator but one the biggest tech firms out there. Not sure how that happened ... people talk, I guess.

So if you don't get in to Y or others I say send a reply making them doubt their decision. This worked for me this past April and more so in 2008 when I was accepted into another popular incubator program.

I'm going to go out on a limb and say that sour grapes probably had no bearing on your further success.

No sour grapes here!

Being courted by one of the biggest tech firms two months into developing our technology was and always be a huge win for my team and I!

> Hmmm I was rejected this past April and then sent a reply back saying something like your loss.

Not exactly an attitude that will impress VC's, and you'd be surprised at how small a world high tech is.

Did you read the whole comment? His whole point is that the email did seem to impress at least someone, and word did get out about his company, and that it was a good thing.

I had zero help and made the mistake of a lifetime with my co-founder. My application to YC became fodder for landfill. After the YC door closed, something miraculous opened. I have now in my lap an opportunity that would be unbelievable even in a movie script. YC is a guidepost. Whether you go or not. The basic resources are all here on HN--excellence and a friend or two.

Great story. I was a bit astonished about the lack of backups as well, but I'm way more astonished that nobody seems to be astonished about the fact that the break up was sent by email and it didn't really contain a reason except for "I don't want to". which is only okay if you're under six years old.

As an adult you're not beholden to anyone to justify your decisions. It's more courteous to explain yourself, but it is not a requirement. I am sure the two of them have discussed it in detail since, but in the moment the important thing was that he was sure.

Are you still at Sift Science?

I'm not (that's a whole separate topic), but the company is still going strong -- if you happen to run a site that accepts payments, and you're dealing with fraud, you can try it here: https://siftscience.com/

Just curious, what are you up to these days? Got another startup cooking, or are you looking for the next place to hang your hat for awhile before you go out and kick ass again?

I have some ideas in-progress, but nothing completely firm yet!

I'm glad it worked out for you! Your product looks solid, and I like the idea. I think one of the key takeaways from this (for me at least) is to have a solid support network.

Thanks! Yeah, part of the reason YC has founders from previous batches hanging out in the main room is to provide a little support to all the nervous interviewees. It definitely calmed me down.

This is a fantastic story. Thanks for sharing.

I like how about 15 of the 30 or so comments right now are about the lack of a backup. This is something that stood out to me too, and I'm happy to see people at least in the tech sector are taking it more seriously.

If you have car/health insurance then paying for backups should make sense to you... if it doesn't your data is clearly not very important (to you).

Just emailed the story to my cofounder to scare him off a little.

Thanks for sharing.

Great post and thanks for being open to what happened. I'd say it's actually a good thing that your would-be-cofounder quitted early thou. I think it'd be worse if he quitted after getting in YC OR right after getting investors' money.

Hey Brandon, I would be really interested in the FAQ questions that you prepared for the interview! Any chance you could publish them (omitting the answers if that's sensitive information)?

Sorry for the delay -- it took me a while to dig up a version. I posted it here: https://docs.google.com/document/d/1hiLQjlCIfBQobLwj2lecohUq...

These are raw, rough notes -- we didn't bother to polish them much since they were only for our own usage. But hopefully they're still useful as an example of how one team prepared for the interview.

Thank you, this was helpful! I appreciate it.

Thanks a lot!

It's surprising to see a person who has rubbed shoulders with the best (At Google and UWash) still feeling nervous before a demo. Could you elaborate on that ?

You never stop feeling nervous when something important is on the line.

Yep. There was a lot on the line and the chances were quite uncertain -- Y Combinator takes, perhaps, 2% of applicants. I've seen them say no to more than a few smart all-Google teams (or all-MIT teams, all-Stanford teams, etc.).

Agreed, there is a lot of sports psychology related stuff written about how sportsmen use nervousness to motivate themselves. How do you not show you are nervous ? Because that is not a good signal is it ? I get nervous for interviews I've prepared decently well for. I don't get nervous when I'm either fully prepared or not prepared at all.

As a former gymnast and YC founder, I have some experience with nerves. The best advice is to prepare as hard as you can, and then relax, knowing that you've already done everything you can to succeed.

Obviously, practicing being put on the spot is also valuable - find ways to do things where you have to deliver even while nervous.

Ex-Googler here. Working at Google is very different than making a presentation representing your own idea. At a big company, assuming a collaborative culture, your teammates are rooting for you to do well and you don't go into a meeting expecting that to be the last word if they don't like your idea. OP had already quit his job, staked his reputation/ego on his demo, and had to speak for himself in a tough room, so I think it's reasonable to feel some level of nervousness.

Great story! btw - chess is a lot more about confidence & psychology than purely intellectual

That depends on who your opponent is. Against inexperienced players, projecting confidence can sometimes make them miss the fact that you made a mistake. However, against strong experienced players they will almost always notice and take advantage of mistakes, confidence and psychology will not factor into it.

Notice, for example, how the top AI players are now pretty much unbeatable by humans. They don't make use of any psychological tricks but play purely rationally -- read up about min-max and decision/game theory for a general framework for how such AI is written.

Yes, obviously you need to be smart to play chess. But on the professional level they are all very smart and capable people. So it's not only about being smart but also understanding who you're facing and how you can crash his mind.

>chess is a lot more about confidence & psychology than purely intellectual

Have you actually played more than a few games of chess?

Actually I did and still am. And I have a 9 years old kid studying & playing 4 times a week. I also met a few professional chess players in my life and talked to them about chess.

That's interesting. In contrast, some people, for example, me, tend toward overconfidence in chess.

Specifically, I'm more likely to lose the first (speed-chess) game of the day, especially if I have not played for a few days, than to lose subsequent games because until I have the cold water of reality splashed in my face, I tend not to put in enough mental effort!

Feeling confident, in other words, makes me relax, which is great at a party, but exactly the wrong thing when I am playing chess (or doing math or debugging a program).

(Specifically, I use the fear of losing or anger at just having lost a game to create enough tension for me to play chess well. By "tension" I refer to activation of my sympathetic nervous system, i.e., what is informally referred to as "adrenalin".)

When I am carrying enough tension to play chess to my potential, I tend to get pessimistic -- with the result that I resign a game every now and then that I should have continued, but that effect is very small compared to the (opposite) effect of my underestimating the amount of mental effort required to win.

I'd be interesting in hearing more about how chess is for you (or your 9-yo kid).

> Six days before the interview, my laptop crashed and we lost our entire demo

ALWAYS use source control!

Source control isn't backup, though a remote repository can be both source control and backup.

    git commit -m "Source control!"
Always store it off-site, too!

Very inspiring. I have no interest in start-ups and that sort of thing, but I do admire your intestinal fortitude. You hung in there and did it. And really that's all life is about.

Really inspiring story. It reminded me - it's time to make backup!

Awesome story, even if it's old!

> Six days before the interview, my laptop crashed and we lost our entire demo.

Wow! Backup much?

really inspiring

- This is a great story, Im happy that you held it through.

> Six days before the interview, my laptop crashed and we lost our entire demo.

Not exactly a ringing endorsement of the qualities of the poster as a software engineer. You'd think that for a project that might change the rest of your life, you'd spend some time reading up about back ups or source control.

This comment adds nothing to the discussion, because absolutely no one reading HN is unaware that that was preventable, and serves no purpose other than to mock someone building things. That should be anathema here.

Not sure I agree. Everyone thinks that backups aren't necessary until something happens to them.

Do you think the author had never heard of svn/backups until it happened to him?

> Everyone thinks that backups aren't necessary until something happens to them.

On the contrary, everybody knows backups are necessary, the problem is bothering to actually do it regularly, especially on a high-intensity time of pre-launch. Personally I forgot to eat at times when working on our Kickstarter campaign video...

Agreed! I'm surprised of the number of comments of this ilk, as if it was the main point (or even a point) of this story.

If the OP had walked onto dog poo just before entering YC's offices and had brought an awful smell to the interview (which would belong to the same category of small catastrophes), would people point out that you really should watch your steps??

Besides, isn't it possible (not necessary, just possible) that successful entrepreneurs are the kind of people who don't have time to backup their stuff, make their bed in the morning, store 10 days of food in the fridge and buy stuff in pairs in case one breaks?

Earlier this year I had 5 computers die on me over a period of a month and a half.

Guess who is a lot more conscientious about backups and source control than he used to be.

With all those new computers I'd say you don't have to worry about backups for at least another 2 years.

I guess that comment was cynical, but just to be sure: http://blog.backblaze.com/2013/11/12/how-long-do-disk-drives.... Especially in the beginning there is a higher failure rate for disk drives. So it is actually the opposite: backups should be one of the first things to arrange for new machines.

> Not exactly a ringing endorsement of the qualities of the poster as a software engineer

Being able to write some code doesn't make you know anything about the rest of the system.

just beware also.. git or subversion can be corrupted. so the best multi server zip/tar backup

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact