Ed McManus, a former technical co-founder of a startup, became a technical advisor for the TV show Silicon Valley. He joined the show after being recommended by his investor and had a unique interview experience. His role on the show involved providing input on technical details and strategy, but not pitching jokes. He helped fill in the technical gaps in the scripts and ensured that the technical details served the story.

- The speaker was involved in writing and reviewing technical dialogue for the show Silicon Valley.
- The challenge was to make the technology in the show fail, even though it was initially portrayed as too good to fail.
- The compression algorithm in the show, called middle out, is based on real machine learning developments applied to compression.
- While the algorithm is plausible, it would be difficult to achieve its performance characteristics in the real world.
- The show reflects the changing dynamics between startups and big companies like Google and Facebook.
- Some episodes were cut and had to be replaced with new episodes due to production reasons.
- The show drew inspiration from real startup experiences, such as launching a beta and facing uncertainty about its success.
- The show also explored the concept of growth hacking, which started as a legitimate strategy but gradually became more questionable.
- The show references real stories and figures from the tech industry, but only if they were already well-known.
- The speaker wanted to make the show more authentic by having the characters use AWS or Heroku instead of building their own servers.

- The speaker was able to convince the show to move to AWS without much pushback from developers.
- The show pays attention to detail in the background elements, such as documents and whiteboards, which the speaker and other consultants work on.
- There was an accidental viral moment when a character's GitHub profile was created and people started searching for it.
- The speaker had ideas for incorporating real software elements into the show, such as encoding text in an audio file.
- The introduction of new characters in the show is mainly for story purposes and not for technical reasons.
- There are three possible routes to get into a position like the speaker's: working hard at startups, being an expert in a specific field, or working at an agency that collaborates with production companies.
- The speaker is not currently considering working on other shows as they prefer working on actual startup projects.
- The shows "Mr. Robot" and "Halt and Catch Fire" are praised for their technical accuracy and attention to detail.
- Silicon Valley has set a high bar for incorporating technical elements into TV shows.

- The trend in Silicon Valley is to produce higher quality episodes with shorter season orders
- Amazon and Netflix are investing billions in new programming and may hire more technical people, even for science fiction shows
- Working on the show was a cathartic experience and helped the speaker get back into coding
- Y Combinator and Hacker News were not well-known or understood by the show's writers, but the speaker thinks they are significant in Silicon Valley
- The speaker suggests that Y Combinator and Hacker News may be mentioned in the next season of the show

- Ed McManus, a former technical co-founder of a startup, became a technical advisor for the TV show Silicon Valley after being recommended by his investor.
- His role on the show involved providing input on technical details and strategy, filling in technical gaps in the scripts, and ensuring the technical details served the story.
- The show reflects the changing dynamics between startups and big companies like Google and Facebook, and draws inspiration from real startup experiences.
- The show pays attention to detail in background elements, such as documents and whiteboards, and incorporates real stories and figures from the tech industry.
- The speaker had ideas for incorporating real software elements into the show and wanted to make it more authentic by having the characters use AWS or Heroku.
- The show sets a high bar for incorporating technical elements into TV shows and the trend in Silicon Valley is to produce higher quality episodes with shorter season orders.
- Working on the show was a cathartic experience for the speaker and helped them get back into coding.
- The speaker suggests that Y Combinator and Hacker News may be mentioned in the next season of the show.

Ed McManus was the technical advisor for season three of Silicon Valley on HBO - https://twitter.com/emcmanus

He came in to talk about what his experience working on the show.

Read the transcript here - http://blog.ycombinator.com/the-technical-advisor-for-silicon-valley-on-hbo-ed-mcmanus

YC's podcast is hosted by Craig Cannon - http://twitter.com/craigcannon

He was a technical advisor for Silicon Valley on HBO, season 3.

Okay, so I was a technical co-founder of a Y Combinator startup called Yard Sale.

And we launched 2 products, both were marketplaces for iPhone.

So similar to Craigslist for iPhone, the first product was called Yard Sale and the second product was Fobo.

And somehow you became a technical advisor for a comedy show.

So, the sort of story here is, we were cranking hard on our startup for 4 years Sold it in 2015 And the acquisition It was kind of mostly just to sort of do right by our investors return as much as we could But it was not like you know phenomenal outcome for the company But all of our investors had sort of like been aware that we we were sort of like looking for our next thing and it just just happened that randomly someone from HBO reached out to Greylock, who was 1 of our investors, and asked them the question, like, hey, we're trying to put together sort of like this team of superstars from the tech world,

And Both me my co-founder were on a list of names that Greylock sent back,

I think it was you know just a matter of being available.

It was vaguely unemployed, sort of funny.

And I mean, it was kind of a crazy story because they sent us an email And I've been a huge fan of the show for a really long time.

And as soon as we got that, it just was apparent to me this is a drop everything and make this happen kind of opportunity.

It was kind of weird, because it actually wasn't really an interview.

So we actually went on separate days, but we both had a similar kind of experience.

I flew in and grabbed lunch with 1 of the other technical advisors from the show, and he's like, okay, you're all set for your interview.

So he drove me over to the studio lot, dropped me off, And I showed up at the the Silicon Valley office, and I you know told them like hey I'm here for the interview and no 1 knew what I was talking about So so I told them who I was and they're like oh, okay I think I think we know where you're supposed to go and so they They lead me into the writers room And everyone's in there so like the whole writing production is on yeah, exactly

right I see Mike judge in the corner And it's just this like weird surreal kind of like I know who you are You have no idea who I am like you know anytime you encounter someone that you consider to be a celebrity.

And it turns out that it was just like no 1 had told them either that I was showing up to interviews.

So it was just like they just gave me a seat and I listened for like 3 and 4 hours, 3 or 4 hours and just like gave input wherever I could

You know it turned out that Most of the job is actually just sort of like sitting and listening And it's very rare that you actually have to speak up and sort of like help them course correct But occasionally they would ask a question, sort of like, you know, can you walk us through the fundraising process, given like the company's at like a seed stage.

And you know, I'd provide some color there around that.

And so your first day ends and like you just leave, you just go home?

Yeah, I flew back up to San Francisco, got a call that night, and they said, like, hey, it went great, we'd like for you to join.

Got 1 of my friends to help me out and sold about 80% of the stuff that I owned just to really commit to this and drove down to Los Angeles on Sunday night.

I got in at like 3 in the morning or something on Monday and had to show up at 9 o'clock the next morning for like my first full day of work.

Okay, so like just drop us into the context.

What's happening in the show at the beginning of season 3?

Yeah, so at this point I think like the first maybe 2 episodes had been written.

It was still very early in the development of the season.

But they had outlines maybe for like the first 5 or so episodes.

And then beyond that, it was kind of like, no 1 really knew where the season was gonna go.

So yeah, when I showed up, I mean, it was like, I was just trying to find my footing.

Like, it's just very, like, to be honest, it was hard.

It was really hard because I've never done anything in entertainment.

I've never worked with people like this before.

Everyone there was just like off the charts brilliant.

So almost everyone in the room had a very solid understanding of how startups work.

Not technical though in the sense that like they couldn't write code.

But 1 of the things I was surprised by was how well the writers understood Silicon Valley and just the material they were covering.

I mean, sort of like a common theme of my time on the show was actually that I didn't have to do a whole lot of correction.

So before every season they actually come up to San Francisco for I think like a week or 2 weeks and meet with a bunch of startups.

So every writer goes through this kind of like crash course to get up to speed.

So by the time they actually start working on the season, they have a pretty good understanding of how things work and what life in startups is actually like.

And so then given that, where do you find yourself Providing value every day at the show

yeah So for the most part it was you know like like I said it was kind of The job was explained to me like it's almost like being a lifeguard, right?

So you kind of sit on the sidelines for a bit.

When you see something going wrong, that's when you actually say something.

So I was on hand to answer any questions and that kind of stuff.

But at the same time, I wanted to recognize that I'm not a comedy person.

That could not be further from my skill set.

So I tried to not pitch jokes or do anything like that.

I just recognize that that's not what I'm good at.

So instead I was thinking through the technical details, sort of like the strategy, like given where the company was at that point in time, what are the kinds of things that they're gonna try and do.

And so were there points where you felt that they were going off course and you had to correct it?

Yeah, I mean there definitely were times where that would happen, but nothing major.

I don't think, there were no huge course corrections as a result of my input.

There were definitely some tweaks here and there, but for the most part, if the writers wanted to do something, we could find a technical explanation to justify them doing it.

Supposedly they can do whatever they want.

We can find a justification for almost any kind of story point.

So the way it would work is they would come up with these scripts and we would...

The first step of the process was to sort of come up with these outlines and from there like they'd hand it off to a single writer who'd go off and turn it into an actual script and they would more or less just leave blank spots.

And so the bulk of my job at first was just kind of like filling in the blanks.

And I thought that that was going to be easy, but it's actually it's fairly, It's not super easy because you have to come up with a dialogue that it's like, okay, you have 2 lines to explain the technical details of what's going on here.

You can't get into the technical details, so you have to keep it fairly vague.

You can't get into the details because folks are essentially like they're not gonna understand Or you don't have enough like words to fit in or

I mean It's a combination of both right even if we could dedicate 30 minutes to like just the technical I'm talking about compression.

That's not why people are interested in the show.

The technical details, another 5 years are not going to be super important.

The stuff that people are going to relate to, it's the characters, it's their own motivations, the emotions, that kind of stuff.

And so what examples did you fill in for?

Like What gaps did you have to fill I mean?

Anytime like 1 of the characters says anything technical that was something that that was that either I wrote or or reviewed So for example if it was something relating to compression sometimes like If it were something that were really specific to an expertise, like I'm not a compression guy, but we have folks who are working with the show who are.

So in that case we would send that out to them, get their feedback, and then sort of like try and bring everything together To make it work for the scene

okay, and so then like going back then Can you talk a little bit about the entire premise of Pied Piper like is that possible is the compression algorithm?

They tried to do at the end of season 1 a thing like what's the deal with that

Yes, so so when I okay, so there's a couple of pieces here, but 1 is that when I first showed up, I was like, this technology's too good.

If this were a real thing, Pied Piper would never fail.

And so the challenge is how do you make them fail?

At the end of every episode, every episode arc, things look great.

It's kind of like the, we actually, it's almost like the anti-entourage, right?

That's how some of them were thinking about the show or like entourage is things like start off horrible And then by the end of the episode they're fantastic Right Silicon Valley is the opposite Yeah Silicon Valley is like things seem like they're going so great like everyone's like on track to being a billionaire and by the end of the episode, the sky is falling.

The challenge was to take this technology that's like can't lose and turn it into something that they keep stumbling over.

Oh, the algorithm at the end of season 1.

The surprising thing with middle out is that there is actually a technical background for it and it is actually the result of a lot of machine learning developments applied to compression.

So The theory behind the tech is actually plausible.

This is a piece of software that could exist.

Most likely it would not have the same kind of characteristics that middle out does in the show.

It's great for streaming, it's lossless, it has all these performance characteristics which are very difficult to achieve in the real world, okay?

So like if this thing existed in the real world it'd be like no question an absolute breakthrough,

Well It would it would just be a scientific breakthrough for something like middle app to exist.

So 1 of the things that I was wondering is like, oftentimes we find, at least around YC, that, you know, like Google or Huli isn't going to immediately Replicate whatever you're working on.

Do you ever bring that up that like the startup versus giant company thing is like Maybe not the most common thing to happen in Silicon Valley.

So a lot of the knowledge that the team is drawing on is actually kind of from the dot com boom.

So like in the 90s, so imagine Hooli is actually Microsoft, right?

Microsoft was like this significant threat.

If Microsoft said they were coming after your market, like watch out.

So maybe a better comparison today would be like Facebook.

Google, we all kind of have all of my experiences with Google being sort of the dominant company.

So now I think it's kind of plausible that we can say, oh, well, the big company's not going to go after this small startup.

But I think that's a new thing, and maybe not can be the case for much longer like you know like the Facebook for example is getting a little bit more aggressive going after smaller spaces and so Yeah, I think that that that is like a legitimate concern.

And yeah, if nothing else and it's it's good for yeah, it's good for driving the story forward,

okay Oh were there any were there any things left on the cutting room floor that you were really excited about that?

so yeah, actually there were I think like 1 or 2 episodes at the end of the season.

So like episode, I think like, I want to say episode 9 or something, where the plan was actually to go to Ephemeral.

So Ephemeral is kind of like Burning Man in the Delta here in Northern California.

AG Yeah, I was on a boat that sank there.

And it sounded like it was going to be incredible.

And That actually got cut more for production reasons than for technical reasons.

But that left the show with this hole in the season and so there was like this mad dash at the end of the Writing period to fill it in with like that like 2 or 3 new episodes, which is like which is like incredibly difficult

So what what episodes were those so I like the luau thing or

yeah, it starts there I think I think the the the the the luau is maybe like episode 7 or something.

And so what we were left with is this hole of a couple of episodes at the end of the season and the challenge is like, okay, so we're kind of screwed here, right?

And I feel like if I pointed to any episodes that I added the most to, it would be those 3.

They ended up pulling directly from our experience with yard sale, launching a beta and sort of like not being totally sure at first, like whether this thing is working or not, and getting a really strong, like positive signal early.

And then, you know, like 4 or 6 months in, you're looking at the numbers and you're thinking like, maybe this thing's not as strong as we thought.

And were you guys, so if I remember correctly, that's also when

the numbers are being faked right so that yeah that happened A little bit more towards the end.

That's like things were getting so dire, they were about to run out of money, that they go off and start faking some numbers.

That originally was going to be sort of a plot around growth hacking, and then it just edged its way into fraud.

But there was this whole fascination around growth hacking and buying users and that kind of stuff.

And of course, take yourself back to 5 years ago in the startup world, 4 or 5 years ago.

It seemed like a legitimate way to get your product to product market fit.

And then it entered this gray hat world and just got gradually worse and worse.

Yeah, and so the show ended up being kind of like a reflection of like sort of like where that was going.

Because I mean like, yeah, there is definitely legitimate and sort of like bad, like growth hacking, right?

And then were there moments like, you know, when you think about like growth hacking in the industry Or whatever they're often like prominent figures Were you ever asked to like put someone on the spot?

And it like felt like 2 on the nose like making fun of them directly or is that just kind of like all fair?

ADAM Yeah, we have a reference to Theranos in the show but that was kind of like, it was very common knowledge at that point.

I think that The show would not go after anyone unless it were already a big story.

A lot of what the writers end up using is stuff that gets pulled from the headlines.

It's real stories that they reshape to fit the situation that the Pied Piper guys are in.

Because that's what I always wondered right like it always feels like not to say that it's like basic but it is like popular culture Silicon Valley right and so were there moments where you were trying to like push it to be like more like ephemeral isle is actually a really good example, but Were there other like more like fringe things people were trying to put in that folks would be like I don't know about this And then take it out

Yeah So like 1 of the I remember watching like season 1 in season 2 and thinking like why are these guys building?

This doesn't make sense No 1 in the startup world does this.

Season 3 comes around, I'm finally in the position to do something about this.

And so 1 of the things that I was like, we have to get the guys to move to AWS.

Like I think that if we could have gotten them to Heroku, that would have been even better.

But there was like, the audience for Silicon Valley is interesting, right?

Because it's not all startup people, but there are a lot of technical people who are really into it.

And so a lot of the technical folks who watch, they're maybe not on the bleeding edge of trends that are catching on in the startup world.

For startups, if you're not building on AWS, You're really behind.

And so watching the show as someone who had actually done a startup like that was something that didn't sit well with me.

So yeah, got them to move to AWS, had to make a case for that.

And we expected a lot of pushback from developers watching the show saying, like, oh, they would never move to AWS, they want control over their own servers, whatever those complaints might be.

It was fine, I think no 1 really complained, so the timing ended up being good.

So what about all those other elements where the show touches the real world?

So we talked before we were starting the interview about you doing like things that were like off-screen, on-screen elements that weren't writing.

Like when do those things come into play and like what did you work on in particular?

Yeah, I mean, so if you watch any episode, just pay attention to the stuff in the background because like the show is designed to be Possible like like you can freeze-frame anything and like look at the documents for example in their hands We're like the whiteboards in the background

all of that stuff Is like me and or like you know some of the other consultants working on that.

Like the scrum board is a good example of that.

Every post-it has something that we had to sit down and spend time thinking about.

I both love and hate the process of making those whiteboards.

There were times where walking the cast through…there's 1 scene in particular during the user feedback session where Richard gets up and starts drawing some stuff on a whiteboard.

So I had to walk him through each of those pieces of, okay, this is what this graphic here means, and that kind of stuff.

There was another time where we had some code displayed on a projector.

And I got to walk Mike through what this code actually did.

I think in the second episode we hid an Easter egg in some C code that we had just flash on screen for a second.

Pausing it, copying it over, and actually compiling it.

I wrote like a really simple compression algorithm,

Like it's hard to even call it a compression algorithm But like I took this this character string and you could you could decompress it expand it into a larger string which just got printed to the console.

I can't remember what message we settled on.

It was just like that attention to detail.

Were you guys on GitHub doing all that stuff?

And so the show has not done a ton of like crossover into the real world.

Last season we had this character Winnie.

And I actually, I created a GitHub profile for playback purposes.

We wanted an authentic looking GitHub profile.

The best way to do that is just to make a GitHub profile for that person.

So I made 1 for Winnie, And we had to populate it both with a fake commit history and also with projects that she would have pushed to.

So set all this up, got the screenshots and the playback finished.

As soon as it aired, people were searching on GitHub for this character's username.

I was just instantly flooded with pull requests.

It was 1 of the first times that the show really crossed into the real world in an authentic way because it was accidental.

AG Were there ever any moments where you were actually thinking about building some kind of software thing that would overlap or was that just never crossing the mind of

I was working on an audio project at the time and realized that we could pretty easily encode text in an audio file.

And so 1 of the pitches in the writer's room was actually to have a character who would play a little clip of audio for like, you know, some some fantasy app that he was he was working on and anyone with the corresponding App could actually sort of decode the audio and it'd be like, you know secret message that We could send out.

Yeah, exactly a little Easter egg That we could send out to all the viewers.

And then what about the people being introduced in the show, the other characters?

Do you as a start-up guy for the show, are you trying to pull out those characters to throw them in or do they never even ask you for that kind of thing?

All of that, very little of that comes out of a technical need.

So it's almost exclusively just for story purposes of moving things forward.

And so There's an opportunity to provide some input and say, well, this is what that person might work on.

And that would be a good example of a blank left in the scripts, so someone's title.

Even the small things like that, which are not necessarily...

It's not related to programming, but it is a technical detail.

And so you can give them a title, for example, that's kind of tailored to the startup world.

And so that's a good example of the kind of thing that

But you don't have to like predict the future and like bring people in and okay that makes sense So I know this is like a very obscure kind of job But if someone who is in the startup world wanted to get into it, how would they get into it, and then how do they get good at it?

See all Y Combinator transcripts on Youtube

The Technical Advisor for Silicon Valley on HBO: Ed McManus