March 31, 2026 • 42 mins
Big tech isn’t a buzzword anymore, it’s the scaffolding holding up the modern economy and, increasingly, modern politics. We sit down and map the real shape of power behind the Magnificent Seven: Apple, Microsoft, Amazon, Alphabet, Meta, NVIDIA and Tesla. We talk through what they do, why they dominate the S&P 500, and the part most people miss, where the revenue comes from versus where the profit actually lands. If you’ve ever wondered why Amazon can run on thin retail margins while AWS prints cash, or why Google Ads is still one of the greatest business models ever built, we make it plain.
From there, we zoom out to the global dependencies that make big tech feel both impressive and fragile. Taiwan’s TSMC sits underneath much of the semiconductor supply chain, and that reality turns “chips” into geopolitics. We also touch on non-US giants like ByteDance and Samsung, then bring it back to the West Coast to ask why Seattle and the Bay Area became such powerful innovation hubs in the first place, from universities and defence roots to talent density and network effects.
Then we get into the part everyone really wants: what it’s like inside these companies. We unpack Silicon Valley compensation and culture, including base salary, bonuses and RSUs, how vesting creates golden handcuffs, and why perks like free food and on-campus services can be both brilliant and slightly manipulative. We also talk about the uncomfortable employee vs contractor divide, and what performance cultures look like when KPIs and reviews are relentless.
Finally, we tackle the looming disruption: AI coding tools like Claude Code, vibe coding demos, and what happens when “writing code” stops being the main job. Are we heading towards fewer engineers, better engineers, or just a different definition of software engineering altogether?
Subscribe, share this with a curious friend, and leave us a review. What part of big tech do you want us to unpack next?
Like, Subscribe, and Follow the Tech Overflow Podcast by visiting this link: https://linktr.ee/Techoverflowpodcast
Listen
Watch
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Hanna (00:03):
Why does it matter if the person gets the code that they need
at a price that they thinkis fair?
Hugh Wil (00:07):
You can vibe code up a demo to demonstrate any feature or
product idea that you have.
So turn up with somethingthat works.
Hannah (00:13):
For someone who doesn't work in that industry, that stuff
really shiny and sexy andappealing.
Hugh Will (00:20):
world, and welcome to the Tech Overflow podcast.
My name is Hugh Williams.
Hannah C (00:24):
And I'm Hannah Clayton Langton, and we are the podcast that
down technology for curiouspeople.
Hugh Willia (00:30):
How do I do Hannah?
Hannah Clayton-L (00:32):
You did great.
Uh that felt weird.
Guys, we decided to do someA-B testing and mix up who does the
intro.
So if we get much morelistens and engagement on this
we'll make that a permanentswap.
Hugh Willi (00:42):
Or we'll try another experiment, one word each, or I
I'm not easier.
Hann (00:46):
And guys, this is a signal that I am now an official tech
and I will be leading thisepisode as our resident technical
Hugh Williams (00:53):
Yeah.
So our topic today, uhHannah, is transistors, the
of the transistor.
So um tell us about transistors.
When were they first invented?
Hannah Clayton-L (01:00):
Okay, correct.
I've been outed pretty quickly.
Okay, we will do an episodeon transistors later on, let's say
season seven.
be by then.
So, no, we're not talkingabout transistors today, guys.
We are talking about big tech.
Hugh Williams (01:15):
Yeah.
Obviously, I spent a lotof my career working in big tech.
So I'm really uh excited,Hannah, to talk about all the
the West Coast tech,maybe a little bit about Taiwan,
and really give our smartlisteners out there some insight
how these companies operate.
Hannah C (01:27):
Guys, my goal for this episode is to get you to spill as
secrets as I can aboutSilicon Valley.
So wish me luck in Endeavour.
Um, I reckon we getstraight into it because I really
mine you for gossip.
no episode about big techwould be complete without a
to the Magnificent Seven.
So those are the sevenbiggest tech companies in the US,
(01:48):
listed, I guess, biggesttech companies in the US.
And I'm sure you can dothis, but I'm gonna ask you to name
seven to kick us off.
Hugh William (01:55):
Yeah, so Apple, of course, uh, Microsoft, Amazon,
which owns Google,amongst other things, also owns
an episode on in season one.
Uh Meta, which of courseowns Facebook, Instagram, and
NVIDIA, which make uh theGPUs that uh we've talked a lot
(02:15):
in season one, and Tesla.
Hannah (02:18):
Ding ding ding, got them all right.
So they make up, I think,about 30% of the SP 500, which is
US's largest 500 publiclylisted companies.
There'll be a few on therethat listeners will know that we
mentioned just becausethey're not publicly listed.
So like OpenAI, Anthropic,right?
They're not.
Hugh Williams (02:34):
Yeah, SpaceX.
I mean, there's there'squite a few large companies that,
know, may publicly listin sometime in the near future.
Uh, but they're the majorones that are publicly listed.
And Magnificent Seven's anew moniker, if you like.
It hasn't been around that long.
We used to use themoniker Fang, which had two A's in
I think is Facebook,Amazon, Apple, Netflix, and Google.
They were the cool kidson the block probably five years
(02:56):
but now we talk about theMagnificent Seven instead.
Hannah Clayton-Lan (02:59):
Interesting.
So Netflix dropped out.
And I'm surprised that youdidn't mention Microsoft.
I would have thought theywould have been in there.
Hugh Wil (03:04):
They were a really big comeback, really, under under Satya
leadership.
And the Azure business isa very, very big business.
They're cloud business.
And we talked about clouda little bit in season one,
in our in our wrap episode.
And also, you know,they've been major investors in AI.
They're a largeshareholder of open AI, famously.
And so, you know, they'vereally made a comeback, and so
(03:24):
part of the Magnificent Seven.
Hannah Cla (03:26):
Interesting that you mentioned cloud because I did a
of research on the revenuestreams of the Magnificent Seven.
And some of it's quiteintuitive and some of it's a bit
So yeah, Microsoft, a lotof the profit driven by cloud.
Speaking of cloud, I wassurprised to learn that I think it's
something like 75 or 80% ofAmazon's revenue is still coming
their retail business, buttheir profit margin is mainly driven
(03:47):
by AWS.
So things like cloud, Iassume, are a much better play than
know, selling bits and bobsonline to consumers, but then the
retail business becomes thevolume play.
Hugh Willi (03:58):
Yeah, yeah, exactly.
I mean, low margin, highvolume business, the retail
And then because of thescale of AWS and the ability that,
know, you can really getscale from that infrastructure, the
margins are much, much,much higher.
And they really are thecompany that largely sits under most
other companies thatyou're using on on your smartphone
on the web.
Han (04:18):
And Google have got a cloud business, but they're, I think
like 90% of the revenuecomes from ads, which I think we're
to talk a little bit moreabout Google in an upcoming episode,
but that's not a hugesurprise given the power that sits
their data collection.
Hugh W (04:31):
One of the best business models, Hannah, of all time is
the Google Ads business.
Han (04:35):
Well, we actually should do a whole episode on that.
Hugh Williams (04:38):
We will.
Hanna (04:38):
And then I was interested to learn that Tesla actually make 10
to 15% of their revenue onlike environmental credits, which I
assume is just sort of away for less environmentally
companies to offset some oftheir emissions, like the sort of
old school car productioncompanies or something like that.
Hugh Willia (04:54):
Yeah, yeah, I think that's the case.
And look, historically,Tesla certainly benefited a lot from
you know governmentcredits, government programs,
kinds of things.
So it's a you know it's acomplex company.
Obviously, it's buildingin its end product something that's
incrediblyenvironmentally friendly, an
that can be used to powerall kinds of things.
Long conversation aboutwhether the actual production of the
car and the batteries isenvironmentally friendly.
(05:15):
I think that's a that's aseparate conversation, but certainly
they've benefited a lotfrom you know regulatory and and
support.
Obviously, a lot ofthat's been wound back by the Trump
Hannah Cl (05:25):
Uh, we won't get into politics overflow on this podcast.
Find us in the pub for thatconversation.
Microsoft's an interestingone because I don't know how widely
known their investment inopen AI is sort of outside of the
world.
And also, I mean, less sexythan ChatGPT, but um, they own
which I only learned when Iwas prepping for this episode.
Hugh Williams (05:43):
Yeah, yeah.
Looking, you know, theyown I think 49% of OpenAI.
And yeah, I think they'veowned LinkedIn since probably around
2016, Hannah.
Hannah Clayton (05:52):
That is correct.
And they bought it.
Can you guess how much for?
Hugh Willia (05:55):
Oh, look, my mind's saying 9 billion, but I bet it's a
bit more than that.
Han (06:00):
It was $26 billion, which I I mean anyway.
Off by an order ofmagnitude, that's me.
Well, I would have guessedmuch lower.
Um, anyway, so yeah, sosome interesting sort of makeup of
revenue of those companies.
You can certainly, youknow, dig a bit deeper.
We won't get into that anyfurther for now.
And then, of course, theMagnificent 7, those are all by
US-based.
We do see a huge amount ofbig tech innovation come from the
(06:23):
coast of the US, and we cantalk about why that is maybe a bit
later on.
But maybe a nod to Taiwanfor the semiconductor production.
So TSMC, I believe thatcompany's called.
And then we've got ByteDance who own TikTok and are
based in China.
More on them later.
And I think it'd probablyfair to call out Samsung in Korea.
Do you think we've missed any?
Hugh Willi (06:43):
Yeah, we've probably missed quite a few.
I mean, there's a lot oflarge Chinese companies, Ali Baba,
Tencent, you know,there's a long, long list.
We should do an episodeon Chinese tech.
And then, you know,there's there's companies that
large as those that arebased outside the US that are pretty
famous.
So, you know, an examplemight be uh Spotify.
Yeah.
Hannah Clayton (06:59):
Yeah, that's it.
I had didn't realise for along time that they were Swedish.
I think that's super cool.
Hugh (07:03):
But on TSMC, I mean, they are the company that really produces
the silicon for most ofthe Magnificent Seven.
So, you know, the supplychain from Taiwan is very, very
to US tech.
And again, that'sprobably a topic we could dig into
season.
Hannah Clayto (07:16):
Yeah, definitely.
The the geopolitics of itsuddenly become quite scary.
Um, but that's not whatwe're gonna focus on today.
I want to take us back tothe west coast of the US because
where Hugh you spent a lotof your career.
But like, why do you thinkwe got these hubs of huge tech
coming out of it?
It's Seattle as well asCalifornia.
I was gonna say maybe itwas the weather, but I think the
(07:36):
in Seattle's a bit crap.
But um, yeah, like why doyou think those hubs ended up there?
Is it just because it'swhere Bill Gates was from?
Hugh Wi (07:42):
You know, the Microsoft story is really the Seattle story,
it's probably not superwell known.
You can read about it inum Bill Gates' latest book, which is
called Source Code.
You know, Microsoft'searly years were really in
New Mexico, and then Billand Paul Allen decided to actually
move the headquartersback to Seattle, which is where they
both were from.
(08:02):
So it was almost anAlbuquerque, New Mexico business
for a while there itbecame the center of tech in the US.
Hann (08:09):
Well, uh Amazon is Seattle too, is that right?
Hugh Willia (08:12):
Yeah, that's right.
That's right.
And and you know, I thinkit's a beneficiary of the fact that
Microsoft was there andreally turned Seattle into a into a
tech destination.
Hannah C (08:19):
Because once you get a critical mass of good engineers, I
is mostly what are indemand, then you want to co-locate
can try and steal that goodtalent, right?
Hugh Williams (08:28):
Yeah, absolutely.
And you know, if you workin any of these companies, you'll
find that there's asignificant number of people who've
there for their wholelife, but most of the other people
come from one of theseother large tech companies.
So there's lots of peopleat Google who are from Microsoft,
lots of people atMicrosoft who are from Google,
know, that these folksjust move around.
Han (08:44):
Okay, so Seattle, that sort of makes sense.
And then we travel down thecoast to California.
Why do you think that built up?
Hugh W (08:51):
Yeah, we fly over Oregon and get to and get to California.
Hann (08:55):
Funny that, not a huge hub of tech, globally, Oregon.
Hugh Williams (08:58):
Yeah.
And if you've ever livedin California or have been to
you know that it's reallytwo states.
So they call it NorthernCalifornia or NORCAL and Southern
or SoCal.
And you know, they'rereally different places.
Um, arguably it could betwo states, you kind of rule a line
across the middle.
And LA is perhaps famousfor the film industry and all those
kinds of things, themusic industry.
It's a very, verydifferent culture to the culture you
(09:19):
find in the Bay Area,which is the area just below San
which is, you know,famously known, I suppose, outside
Bay Area as the Silicon Valley.
So you'll find the peopleinside the so-called Silicon Valley
just call it the BayArea, and uh the rest of the world
it the Silicon Valley.
But I think the historyreally is that we had Stanford
obviously one of the youknow, world-class university.
(09:41):
It's always been very,very good at computer science, and
it's always produced topshelf graduates.
Larry and and Sergei, whofounded Google, are, you know,
folks.
So lots of lots of famouspeople have come out of Stanford and
gone on to foundcompanies, but it's historically
very strong school forcomputer science.
Uh, defense was alsobased sort of in that region.
(10:01):
So there's lots ofdefense industry there.
Climate's great.
Back in the day, therewas plenty of land, lots of fruit
And so, you know, you puttogether the universities with the
defense investment, witha wonderful climate, plenty of land.
And the end result is itbecame really the center of tech in
the world.
Han (10:17):
Okay, so it literally comes down to accessibility of smart
I know from having workedwith robotics through my current job
that there's like a big hubaround Boston for similar reasons.
You've got like Harvard andTufts and Boston College and some
of the really good unis.
So it sort of makes sensethat that's where the employers
to.
And what years were you in?
You were like early SiliconValley, right?
(10:38):
Is that fair?
Hugh W (10:39):
No, no, no, no, not fair at all.
No, no, Silicon, no, I'mnot fairly.
Maybe we can take that out.
Hannah Clayton- (10:43):
We'll let's go.
Hugh (10:44):
No, I'm definitely the uh senior member of the team here, but
I'm not quite that senior.
Hannah (10:48):
Okay, so what years were you in Silicon Valley, aka the Bay
Hugh Wi (10:52):
Yeah, so um we moved to the Bay Area in 2009, and I was
until 2017.
So just a small part ofthe history of uh of the so-called
Valley.
Hanna (11:04):
But that was a really big so maybe just since you've mentioned
timelines, just to placesort of what was going on in the
of tech, because I I foundthis quite interesting as it came
of my research.
if we think about thehistory of what tech meant, right?
Because back in the day, soin the 1980s, let's say it was all
about personal computing,the internet wasn't really a thing
or not something that theeveryday person was using.
(11:24):
And then the 90s was allabout the internet boom.
I I literally remember thefirst time we crowded around my
desk, um, actually also inthe US, and dialed up to the
with that very distinctivesound.
If you know, you know.
2000 to 2010 was all aboutsocial media.
And then we moved into appsand the app economy and app for
And now when we talk abouttech, we're really talking about
(11:46):
cloud, generative AI,platform tech.
So through that timeperiod, tech as an industry meant
things depending on whatwas going on.
So you were there duringthe sort of social media, mobile,
app economy phase of things.
Hugh Williams (11:59):
Yeah, that's it.
And probably the birth ofcloud computing, really, as it
something that becamecommoditized and available to other
around the world.
And we've spoken a littlebit about it in earlier episodes,
Hannah, where we'vetalked about sort of how Microsoft
its own cloud for its ownpurposes, and in its case for Bing,
and Google did the samefor its Google search, and Amazon
the same for its store.
And then obviously theywent on and sold those things to
(12:20):
people, and that's reallynow what powers the internet.
But certainly I was therefor the the mobile revolution, maybe
some part of the webrevolution and definitely the cloud
Hannah (12:29):
And does that change the type of people like computer
that you need?
Do you need differentskills or is it all covered in a
science degree?
Hugh Williams (12:38):
Yeah, really good question, Hannah.
I mean, if you zoom out alot, right, I think at the beginning
of computing, it was allabout hardware.
And you know, softwarewas a small part.
You know, certainly overtime, what's happened is hardware's
become, you know, it'sstill interesting, but the the real
has gone into software.
And look, I think a goodcomputer science degree gives you
problem solving skillsthat you can use to adapt to any new
(12:58):
technology.
So that's one of thewonderful things about doing
is you learn how to be aproblem solver.
Um, you learn how todecompose a problem down into
and solve those problems.
You know, we talked alittle bit about that way back in
first episode of season one.
And so, you know, atleast on paper, I guess uh somebody
a computer science degreeought to be able to adapt to changes
in technology.
But I say on a morepractical level, the kinds of skills
(13:20):
are valuable has haschanged quite a lot.
I mean, there wascertainly a phase when maybe we'll
back to the start of theweb where being a web developer,
could build web pages andmake those interactive was kind of
the the hot skill.
And then being an appdeveloper became a very valuable
You know, if you couldwrite iPhone apps or Android apps,
was a valuable skill.
Again, that's a littlebit more commoditized now,
(13:41):
with things like ClaudeCode, where you can generate large
of code.
And these are often verysimple apps to build.
Then we sort of wentthrough a phase of where the sort of
infrastructure people whocould stitch together AWS and Azure
and Google's cloud toreally build scalable systems were
valuable.
Those folks are stillquite valuable.
But what I think whatwe've really seen over the last few
(14:02):
is the emergence of thedata scientists.
So these are people whocan use data to create products, and
they're often people whoare very skilled at machine learning
and AI.
And obviously that is theuh topic de jour, if you like, of
uh of computer science atthe moment.
Hannah (14:16):
Yeah, I wonder how that will develop.
If you think about it, likeif someone works for 40, 45 years
in their career, that'sspanning a lot of different eras
tech.
And so I guess you eitherhave to be, you know, super
to the thing that you do,even if it becomes out of fashion
sort of the main, or youhave to be pretty agile, right?
And like, would people retrain?
Hugh Williams (14:35):
Yeah, absolutely.
I mean, the programminglanguages that I learned at
are largely not used anymore.
So certainly, you know,through my career, I've had to adapt
to different programminglanguages, different ways of
software, um, you know,pretty much everything's changed.
But again, if you've ifyou've got those fundamental
science characteristics,you should ought to be able to, you
know, adapt to change over time.
And that's what I alwayssay to people when we're hiring, we
should be hiring forfundamental sort of competencies, if
(14:58):
like.
Is a person a goodproblem solver?
Are they smart?
Are they driven?
They got a computerscience degree, not hiring for
who has a specific skillbecause you know, skills come and go
very, very quickly in uhin the tech industry.
Hannah Cla (15:10):
these highly skilled problem solvers that you're drawing
top universities on thewest coast of the US, they're
paid a lot of money, right?
Like if you're working intech, if you're an engineer, there's
sort of two paths you cango down.
Maybe this is my outsideinterview, and you can go for like
lifestyle, work remotelyfrom a beach, get paid pretty good
considering that you'resort of in control and have the
or you can go after like acold, hard cash and land a job in
(15:32):
Silicon Valley and make alot of money.
Hugh Williams (15:35):
Yeah, that's it.
I mean, I think there'slots of things you can do in tech,
certainly if you want tobe amongst the best of the best and
be paid the largestamount possible, then you know,
the Silicon Valley is theis the way to go.
Hannah C (15:45):
And it's not just like monetary compensation, like Netflix
have unlimited holiday.
And I know that Google umsupport female employees freezing
eggs, which is actually areally awesome benefit that I have a
couple of friends that havetaken them up on it.
Free food.
I remember I once worked atYouTube early on in my career.
I wasn't working forYouTube, but we did an event with
(16:05):
and it was like going intothe Google offices and seeing that
they do actually have thenap pods and the massage room and
free food that you can justtake from the canteen.
Like for someone whodoesn't work in that industry, that
was like really shiny andsexy and appealing.
Hugh W (16:19):
Yeah, and look at all of it's true.
Certainly when I workedat Google, I worked at the Mountain
View campus, so the maincampus of Google.
I think it probably hadaround 50 different restaurants.
Wow.
Um, not all of them openfor all meals, but you know, they
breakfast, lunch, and dinner.
Pretty much select anycuisine you want any time of the
So I used to hit the gym.
They they probably had, Idon't know, more than 25 gyms.
(16:40):
They'd sort of done afaux athletics track with the
grass and things.
They had big tractortires there that you could flip
that you could hit thingswith, put out sort of the cones and
have running drills, youknow, a bit sort of boot camp kind
of style.
I used to enjoy that.
And when I finished theclass, go and have a shower and uh
the free towels, ofcourse, uh that come with the uh the
(17:01):
And then I'd head over toone of the smoothie stations and uh
get them to make me up areally, really good protein smoothie
with all the additivesthat that I wanted.
Usually for lunch, youknow, I'd pick one of the healthier
quite like one of theJapanese restaurants they had, used
go there.
And then I didn't usuallystick around for dinner, I'd jump
on my bike and pedal home.
But certainly, you know,the culture is designed to get you
in there early so you geta free breakfast, feed you at
(17:23):
and then get you to staylate, have dinner, stick around,
do a little bit more work.
So, you know, the cultureis designed to get you to work hard
by providing you with theuh the perks to help you do that.
Hannah (17:34):
Okay, I would be such a sucker to that culture.
Like, if you give me a goodgym on, I'll be there on the
Like that is very smart.
I'll take us for a minute alittle bit further away
But my brother worked forbite dance in China for a while.
Oh, wow.
And um I asked him for someof the sort of insights into how it
was working there becausethey worked him pretty hard.
(17:54):
I think he was there fortwo or three years.
And he said similar things.
So it was like core workinghours were 10 till seven, I think.
But free breakfast was fromnine and free dinner was from seven.
Hugh Wil (18:04):
So it's like nine nine six is they call it him.
Hannah Cla (18:06):
Yes, yes, that's it.
Sorry, I should have openedwith that.
Beijing tech is the ninenine six, so you work nine till nine
six days a week.
Yeah, yeah.
Um, actually, sorry, juston the six days a week thing.
He said that when he movedto work at Byte Dance, they were
changing from working sixday weeks every week to working a
week and then a five-dayweek the following week.
And they did somethingreally cheeky where they were like,
(18:26):
we were actually paiddouble for Saturday, so you're gonna
you know, more thanone-sixth of your pay cut on the
you are only doing five days.
And he said that somepeople really didn't like that
just actually wanted towork super hard.
He said there was a ton ofyoung people who worked there
until they burnt out.
And then he said they gotsubsidized rent, but only as long as
it was within 20 minutes ofthe office or something.
(18:47):
So he didn't go for that one.
He actually said that itinflated the rents in the local area
and so it wasn't reallyworth it.
But it was that wholedesign of like, it sounds really
someone like me who wants aconvenient gym will be like
bought in.
But it's basically designedto keep you on site and working for
as much of the day aspossible, right?
Hugh Williams (19:06):
Yeah, absolutely.
And going back to theGoogle story, I mean, they had
much had every serviceyou could possibly think of, and
were free.
So they had like a bikerepair store.
You could, you could, youknow, if you had an issue with your
bike, you could take yourbike there and they'd repair it for
you while you're at work,laundry, all these kinds of things.
Dry cleaning?
Uh yeah, yeah, certainlydry cleaning.
Um, I can't rememberwhether it was free or not, but
they had, you know, drycleaning where they'd come and pick
(19:26):
it up and drop it back.
Um, and you know,probably subsidised at the very,
But certainly, you know,every service you needed to
live your life wasavailable on the campus again to get
to stay on the campus.
The only thing theydidn't have was beds.
So you couldn't actuallysleep overnight there.
Hannah Clayton-Langton (19:41):
Okay.
And did they have likechild care and or pet care?
Like sometimes you see thatlike doggy day care is uh in London,
like dog-friendly officesis a big pull for certain people.
Hugh Wi (19:50):
Yeah, so certainly, you know, all of those things are
of.
So if you sort of zoomout a little bit and think about
of the broader benefitspackage for somebody working at
I mean, certainly theywant you and your family to be
So the health benefitsare fantastic, you know, and a lot
things like, you know,looking after your pet, childcare,
kinds of things are allpretty heavily supported.
And obviously, becauseit's one of the most wealthy places
the world, there's anenormous number of workers who come
(20:13):
that area to providethose kinds of services.
So it's definitely thecenter of the universe when it comes
to sort of being lookedafter, both within the company and
you know, broadly as afamily that lives there.
Hanna (20:24):
you get the same perks if you're like a receptionist or like
a finance manager?
Do you get the same perksas the engineers?
Hugh Williams (20:30):
Yes and no.
I think um there'sdifferent colored badges.
And so you know, theemployees get a get a badge that's
colour and thecontractors get a badge that's a
and you you get virtuallyno perks if you are a contractor.
And you would find, forexample, that pretty much all the
are going to becontractors, contracted through a
And so they're wonderfulpeople, they work hard, you know,
share the ethos, but theydon't get any of the benefits really
(20:53):
that the employees get.
Hannah (20:55):
I had heard that Google, because Google got big offices in
that they mainly hirepeople at least to start on
And I had understood thatthat was because it's easier to sort
of end the relationshipwith the employee if it's not gone
and then they can just makeyou permanent if it does.
But I hadn't consideredthat's probably also a benefits
Hugh Wi (21:12):
Yeah, look, it probably is.
And and look, you know,California has um what we call
deployment.
And that basically meansthat they can get rid of you or you
can quit and you can dothat fairly quickly.
Obviously, places likethe UK and Australia, it's a lot
difficult to get rid ofemployees.
Uh, and you know,sometimes, particularly in the UK,
an employee to leave.
So, you know, some of thesenior executives in UK tech have
(21:34):
to give crazy noticeperiods that, you know, like 12
right?
Yeah, just completelyblow my mind.
Where, you know, the Ithink the longest I've probably ever
seen in the US West Coastis probably three weeks.
Hannah Clayton-Langton (21:43):
Oh, wow.
Okay.
Wow.
Even UK, the shortestyou'll get is one month.
And back to these differentcolored badges because I'm kind of
shook by this.
Okay, so does that createlike a two tier Diddy engineers who
I presume are the people onthe permanent contracts, like, does
that make them sort of likefirst class?
Citizens and then the otherones like maybe less so.
Hugh Williams (22:03):
Yeah, I think uncomfortably so.
I think uncomfortably so.
Hann (22:06):
Okay, so we've kind of got a two-tier system.
And so would your likein-house lawyers be contractors or
they be full-time?
Hugh Wi (22:12):
No, they'd be they'd be full-time.
They though things likepatent lawyers, you know, the folks
who um who write up thepatent applications and things would
be external legal firms.
So when it comes to sortof some of the specialist legal
you know, they use thesewell-known external firms, but
all the professionalstaff who are paid well are
Hannah Clayton-Langton (22:31):
Okay.
And it's obviously pairedwith like caning the working hours,
right?
Like I imagine if you are avery highly paid engineer at one of
the big tech firms, theexpectations on you are pretty high.
And what you've said aboutour well employment means that that
can all be ended prettyquickly by an employer if they don't
like they're getting thevalue from you.
Hugh (22:48):
Yeah, these are these are performance cultures for sure.
Um, and I think most ofthese companies are by any measure
class at setting, youknow, KPIs or performance
of the people that work there.
They by and large havepretty strong review cultures.
So, you know, you sitdown and you talk about your
They have very strongbonus cultures where they'll pay
(23:09):
or not pay bonuses basedon people's performance.
And so certainly theseare cultures where, you know, people
are encouraged to workhard towards very specific goals.
And, you know, if youdon't hit those goals, it might not
you bonus.
And uh, you know,probably not far behind that is you
be shown the door.
Hannah (23:24):
Okay, and talking about bonuses, like what percent of
base would they then bestriving for as their bonus?
Hugh Willia (23:31):
Yeah, so look at it depends on level.
If you're a recentgraduate and you you join these
you know, your bonustarget might be anything from say
to 10 or 0 to 20% of yourof your base salary, which, you
sounds pretty good tomost people.
Once you get up into thesenior executive ranks, you know,
probably in the range ofzero to a hundred percent, maybe
to two hundred percent ofyour base salary.
(23:51):
And so, you know, youstart to think of the bonus as the
part of your cashcompensation, though a very
of your cash compensationbecause it doesn't just depend on
your personalperformance, but it also depends on
performance.
And obviously, you'resomebody who might have some
on that, but you don'tcompletely control that.
Hannah C (24:08):
Okay, but that kind of makes sense at an executive level
the bonus targets would bethat high, right?
And that's not specific totech necessarily.
But what about like asenior engineer?
Like they need to beinvested in what they're building,
way that we do that is bygiving them shares.
That's another part of compwe've not really mentioned shares
or bonus.
Like it do you find thatthat that mid-range gets pretty well
(24:29):
looked after in that senseas well to get them bought into what
they're doing?
Hugh Williams (24:33):
Yeah, absolutely.
And you know, salaries goup very, very quickly in in US tech
from uh junior salarieswhere, you know, you're an engineer
and you're perhapsstarting your first role at a place
a TikTok, you might begetting a base salary of, say,
US, which sounds good byany international standards.
And then once you're inthe door, if you're somebody who
hits those targets, keepsgetting bonuses, I mean, you could
(24:54):
certainly over a periodof years be, you know, getting paid
up around a milliondollars to be an individual
you know, somebody who'ssitting there writing software every
day.
Hannah (25:04):
Yeah, guys, I picked the wrong career path.
Well, who knows wherepodcasting will take me, but
feel I went to an all-girlsschool and it would we were sort of
pushed if you were intoscience that you'd do medicine.
And if not, don't botherwith science.
And there was a big gapwhen it came to things like
computer engineering, maybeit would have been called.
And I think my schoolmissed a trick.
Hugh William (25:22):
Yeah, I think most schools miss a trick, Hannah.
I mean, we should talkabout our charity code for schools
some point and what we'retrying to do here in Australia to
change that trajectory,particularly for girls.
But uh, yeah, lots ofschools don't understand how
tech is and uh how muchyou know one engineer can change the
world.
I mean, you really canwork on a product that affects
of people, and andthere's really very few industries
(25:43):
that's the case.
And so, you know, it's ait's a pretty interesting career to
get into, and because ofthat, leverage that impact, you can
you can make lots ofmoney for yourself too.
Hann (25:50):
Well, think about Jonathan Medine, right?
We had him in episodefive-ish of season one, and he just
one night coding, inventedthe swipe right, and that's
changed the face of moderndating.
So it's definitely not anover-exaggeration to talk about that
impact.
And I have many a timeconsidered retraining.
Think about it all the timethis episode too.
Giving me some more pausefor thought on that as well.
(26:12):
And I think just whilewe're on the topic of the sort of
split, which we we don'tneed to get into in too much detail.
But one thing I have heardis that the female engineers that do
exist, because they're sortof famously, um, it's more of a male
skewed industry, but theycan demand some pretty big salaries
because the big companieshave quotas or they want more female
engineering presence.
And so if you're a strongfemale engineer, you can sort of go
(26:33):
in and demand your priceand they'll be really fighting for
Hugh Wi (26:36):
You know, certainly all the organizations that I've worked
and this concludes inrecent times, they're probably at
90% male and you know,maybe around 10% female, and that's
know, clearly that's notrepresentative of, you know, how the
world is.
And so I think morediversity would be fabulous in tech,
certainly uh, you know,you have to encourage women to get
tech, and then uh thosewomen who get into tech and who are
great at tech, you know,everybody wants them quite
Hannah Clayt (26:59):
Yeah, fair enough.
Okay, girl power.
So back to shares.
Does that not dilute theshareholding of a company if they're
just giving loads of freeshares out to all their employees,
is it not quite as simpleas that?
Hugh Wi (27:12):
It does create dilution in the in the shares, but probably
to a significant extentunless you're really talking about
who are getting, youknow, big parts of of the company.
But it's a very, veryimportant part of the compensation
anybody in the US.
And there's really threecomponents to the comp in West Coast
Tech, Hannah.
And, you know, even youngengineers are pretty savvy about
So, you know, first ofall, there's your base salary.
(27:33):
That's the cash you getevery bi-weekly, as they'd say, in
US.
The bonus.
We talked a little bitabout that.
Young engineers will beinterested in what's the target
and what do I have to doto get it?
Do you ever pay more thanthe target bonus?
Those kinds of things.
And then there's theRSUs, the restricted stock units,
if you like, of the company.
And most engineers willbuild a little spreadsheet and
say, good, better, best.
(27:54):
They'll put in somepossible stock prices, share prices
might be reached.
They'll build out thevesting schedule of those shares and
figure out how much theycould be potentially worth.
And that's certainly areally big factor in the
process of engineers isyou know, do they believe in the
prospects of the companyand will they make a lot of money
of their shares?
Hanna (28:11):
And and RSUs, or at least my experience with them, was that
vest over a certain numberof years.
So it's like, oh, great,we've given you this huge slug of
but you've got to stay herefour more years to get them all.
Hugh Williams (28:21):
Yeah, absolutely.
It's a little bit morecomplex than that, too.
So let's take a fictionalexample.
So imagine you uh youjoin a company and they give you
RSUs as a as an awardwhen you join.
So they say, here you go,you can have 10,000 shares.
What'll typically happenis that they'll vest out over a
period.
There's often sort of acliff before you get the first part.
So it might be say, hey,you have to be here six months
(28:43):
you're gonna get any, andthen you'll get six months' worth.
And then maybe they'llvest every month after that until
reach the four years andyou've got all 10,000.
But then what'll happenevery year usually in these
companies is they'll giveyou another share award.
So they'll say, Hey,Hannah, you've been fantastic.
We're gonna give you14,000 shares this year.
And again, they'll vestout over four years.
(29:05):
And so now you've got twolots of vesting going on.
You've got the awards yougot on the first day you joined, and
then you've got someaward that you got on the 365th day,
anniversary of a joining.
And then guess whathappens next year?
We'll give you another pile.
And so you've got allthese things vesting on different
different numbers of shares.
Han (29:22):
Different share price, like share price fluctuations, right?
Because if there was a youalways want shares on a down year if
you believe in the company,because that's where you're going
to make most of your money.
Hugh Williams (29:31):
Absolutely.
So, yeah, often whatthey'll do is, as you say, is
well, we want to giveHannah a a certain dollar value
of shares.
And so they'll dividethat by the current share price to
the whether it's 10,000shares or 12,000 shares.
And so you're factoringin, you know, the future prospects
the company, the numberof shares you might get, but you've
got this very complexequation going on in your head.
You know, they call itthe golden handcuffs, because
(29:52):
you leave, you lose allof those shares that are yet to
Han (29:56):
And arguably the best share play is to go for like a super early
phase startup that you justbelieve in, right?
Like I had a an oldcolleague once who he, that was his
thing.
Like we were on a gradscheme together, and then he was
My next move, I'm gonna goearly phase, be the single digit
number, and then I'm gonnamake a ton of money once that goes
big.
But basically, if you'rereally good at spotting, you know,
(30:17):
next unicorn, that could,you know, make you could make your
by the time you're 30.
But I imagine that there'smore people who don't than do who
that approach.
Hugh Williams (30:25):
Absolutely.
And different, you know,it's just about different levels of
risk appetite, right?
So there's certain peoplewho are happy to go for that hero
or zero company where,you know, it's either gonna be worth
nothing at all, or maybeyou'll make a fortune.
And there's a certain setof people, and I probably fall into
that category of who arehappy to work for big tech.
You get a generous basesalary, you get a predictable bonus,
you get some RSUs orstock units that you know are going
(30:46):
be worth, you know,roughly in this kind of range,
something special happened.
And something special hashappened over these last few years
as share prices of theMagnificent 7 have run up
Hann (30:56):
All of this must amount to quite a distinct culture in Silicon
Valley, and I'm sure inother sort of tech hubs that exist
in the world.
But like it sounds likethese people might be a little bit,
don't know, like maybearrogant, which isn't a word I would
associate with likeengineering, you know, sort of
but it's kind of crazy thatthey would be making all this money
(31:18):
and be in this bubble whereit's normal to get all this sort of
compensation and it's quitedistinct from the real world that
most of us live in.
Hugh Williams (31:26):
Yeah, yeah, yeah.
And I recommend the theTV show Silicon Valley if you kind
want to see a caricatureversion of this.
It's uh it's pretty funny.
But yeah, certainly, Imean, you know, when I was a kid, I
just kind of some nerdy geek.
And then when you move toSilicon Valley, you know, you're
of a population of peoplelike you, and you know, you feel a
little bit like a rockstar given the money that you're
paid and where you liveand all the things that you can do
(31:48):
that kind of money.
Hannah (31:49):
Okay, so this episode is now sounding like a pretty strong
for people becomingsoftware engineers.
about the disruption ofsomething like Claude Code?
Like suddenly you've got amuch cheaper way of generating that
output.
Even if you don't removeall of the engineers, you remove 50%
of them, 70% of them, thenit's much cheaper to use that.
And then your big ticketsalaries and compensation and
(32:10):
packages might just disappear.
Hugh Willi (32:13):
Look, I think that's partly true and partly not true.
And I think it's a littlebit unclear exactly where this is
all headed.
But if we if we wound theclock back, say six months, I had
the very strong view thatyou know, things like clawed code
were kind of generating,you know, code that could be used in
prototypes to demonstratea concept, but it really wasn't
beyond that.
So everybody's quoteunquote vibe coding up these apps
(32:35):
saying, hey, softwareengineering isn't too hard.
And I'm like, try gettingthat running at scale where you have
99,000 queries a second.
Hannah Clayton-Langton (32:42):
Okay.
And then six months ago,you you're thinking, eh, like good
prototyping.
Yeah.
Hugh W (32:47):
And good for prototyping in a super cool way.
Like we spoke aboutproduct management back in season
think it was our second episode.
I think even six monthsago, you know, my advice to young
managers is don't turn upwith slides, don't turn up with
don't turn up withPowerPoints, turn up with a demo.
Like you can, even ifyou're not a really proficient
you can vibe code up ademo to demonstrate any feature or
(33:08):
product idea that you have.
So turn up with somethingthat works.
Hannah Clayton-Lan (33:11):
I love that.
That's really cool.
Hugh Williams (33:12):
Yeah, yeah.
And I think we shouldexpect that of product managers
It's like go build it.
If you think it's real,build it and show me the demo.
There's even a course nowuh at CMU, Carnegie Mellon
called Vibe Founding.
Hannah (33:23):
What and what does that mean?
Hugh (33:24):
So the idea is, you know, you have a startup idea, and within
a week or two of runningin this class, you actually come out
with a product that'sactually launched and running and
a business model.
Hannah Clayton-Lan (33:33):
I love that.
That sounds way moreinteresting than the medieval French
that I studied in university.
That sounds great.
And so I think you'rebuilding to telling me that since
ago, your attitude changedon that, right?
Like I heard in the roundthat one of the latest Claude models
has really changed the gamein terms of the quality of the
that the LLMs are giving.
Hugh Will (33:52):
Yeah, it's changed so much and in a way that I just did
predict.
So I'm starting to think,Hannah, that coding by hand may be
a thing of the past.
We may not need to dothat anymore.
That's big.
Yeah, and certainly I'mstruggling to find cases where
like clawed code can't dothe work.
I even heard from afriend the other day, maybe this is
(34:13):
down the down the mill,but that Microsoft is now mandated
every software engineerneeds to use clawed code who works
Microsoft.
Hann (34:21):
Well, I mean, if it's that good and it enhances productivity,
that makes sense.
But then is this sort ofdarker undertone as that means you
fewer engineers, right?
Hugh W (34:30):
Yeah, well, the question is what's software engineering,
And I think I like analogies.
We use a lot of thosethroughout the show.
But what I would say isit's a little bit like power tools
the carpenter.
The average carpenter'sgot a nail gun, you know, they've
an electric saw, thesekinds of things.
They plug batteries intothings, things are just a lot faster
and a lot easier, andthey can do a more professional job.
And I would say thingslike Claude Code are a little bit
(34:51):
power tools for the carpenter.
But I'd say that softwareengineering is even broader than
It's more like buildingsoftware.
Hannah Clayton-Lan (34:56):
Yes, we did.
Yeah, we talked a littleabout this, like the architecture,
plans.
Hugh Williams (35:00):
Yeah.
So I think do you needsoftware engineers?
Absolutely, you do.
But will they be coding as much?
Probably not.
And so then I guess thequestion becomes do we need as many
software engineers giventhey spent so much time coding?
So I think I think it'syet to play out.
I think probably theanswer is that the best software
are going to be paid evenmore and uh going to be very, very
valuable.
But I think that bandthat's not so great today, I think
(35:23):
know, could be in alittle bit of trouble.
Hannah Cla (35:24):
And doesn't what you described potentially blow the line
product management andengineering even more?
Hugh Williams (35:31):
Yeah, absolutely.
And I think if I wasgiving advice to a young person
I'd say become a productmanager who understands data science
and vibe code like crazy.
Hannah Clayton-Langton (35:39):
Yeah.
Before LLMs entered thescene, there was a lot of companies
were like offshoringengineering capability to low-cost
I think like India is themain one that I know of.
I actually have a friend,actually, he's a listener who is a
engineer and he works witha team in India and he has teams in
the UK, and he says hegives directs different types of
(36:00):
the teams because theIndian engineers have like less
but they can just churnstuff out really quickly, which is
helpful.
But do you see that beingdisrupted by the LLMs?
Hugh Will (36:10):
I think massively so.
These folks who work, youknow, in these remote geographies
given really specifictasks really are mostly just coding.
Hannah Clayton-Langton (36:17):
Yeah.
Hugh Williams (36:17):
Right.
So they're not reallydoing the whole software engineering
They're just asked to fixthis piece of code, write this
whatever else it is.
And that can prettyeasily be done by tools like clawed
today.
I think that'sdramatically going to change.
Hannah Clayton-Langton (36:31):
Yeah.
Just to be clear, it's notbecause they can't do that software
engineering piece.
It's just that the way techcompanies have set up their offshore
arms is that they give veryspecific like to-do lists to people
who aren't as embedded inthe product.
And so, yeah, it soundslike that's that's the thing that
end up pretty disrupted.
Or in I mean, I guess youcould say it's enhanced, right?
You'll get a lot moreproductivity, but it just means that
(36:51):
will be fewer jobs togenerate the same amount of output.
Hugh Will (36:54):
Yeah, or there's more companies created, more happening in
and those people aredeployed into those, into those new
and new spaces.
Hannah Clay (37:03):
So we'll we'll see.
Tell you what though, youhear these stories where I think,
I think I know some people,this has actually been true.
People have more than one job.
So if you're in one ofthese sort of remote engineering
there's like a good historyof it's probably not just
but it's probably an easyone to sort of fake that they'll
more than one job andthey'll just be like pretty
and sort of be able tofudge doing two different jobs at
(37:25):
time.
And uh yeah, I guess onceyou have this enhanced productivity
expectation, it might beharder to fudge that through.
Hugh Williams (37:32):
Yeah, absolutely.
And I'm I'm sure there'salready people uh lying in hammocks
across the world withlots of agents running, doing all
coding for them andsending bills to people.
Hann (37:40):
Yeah, I mean, I think it's kind of fair enough if you can get
away with it.
Maybe that's a bit morallygray, but like at work, my line is
kind of like if I can'ttell that you've used Chat GPT or
similar, then I actuallydon't really mind.
It's just that when theoutput is it, what is it, the M-dash
that you is a real give.
Like if you can tellsomeone's use it, then they've not
well enough.
But if you can build codeand bill people for it in a way
(38:01):
you know, you're gettinggreat return on your time in the
I think that's fair enoughuntil, you know, why does it matter
if the person gets the codethat they need at a price that they
think is fair?
Hugh Williams (38:09):
Yeah, absolutely.
And I think, look, LLMsare great for two things.
You know, one is to sortof stimulate the thinking in the
place before you actuallydo the thinking and then get on with
the task.
Or, you know, if you'vedone all the thinking, you've
it through, and you justwant to, you know, correct the
write that as aparagraph, turn those bullets into a
whatever it is, then, youknow, power to you.
Hannah Clay (38:28):
Yeah, power to you.
Okay.
And I feel like, is it fairto say that the cadence of change
in Silicon Valley or bigtech in general is just getting
and quicker as we gobecause the world just moves faster
Hugh Williams (38:40):
Yeah, I think so.
I mean, I think uh, youknow, AI is super disruptive, LLMs
super disruptive.
And so there's certainlybeen unprecedented change in the
few years after what Ithink was a period of lull.
But if you look back farenough, you know, we had mobile, the
smartphone, that was apretty big revolution.
We had the personalcomputer was a pretty big
So cloud computing wasarguably as big a revolution, though
(39:01):
a little bit slower on the boil.
But certainly we'vealways had periods of revolution
by periods of stability,but we're certainly going through
of those periods ofrevolution right now.
Hannah Clayton-Lan (39:09):
Interesting.
So don't retrain as asoftware engineer just yet.
Get into data science.
Um, okay, so let's justtake a big step back out to round
this episode, which is whydoes big tech matter?
you're not interested inbeing a part of this industry, then
it just sounds like a bunchof young people making more money
than they have sense todeal with.
But um, there's some prettysignificant geopolitical impacts
(39:30):
days from big tech, right?
So the obvious one is ifTaiwan's invaded by China, that
we mentioned up front, youknow, that could cause some pretty
significant disruption totech supply chains, which will
everyone's life, by theway, guys, because everything is
by tech now.
But also just things likethe latest US administration's
to, well, basically everything.
But I've heard some sort ofEuropean tech leaders casting some
(39:54):
doubt over whether theyshould be using the big cloud
providers out in the USbecause they just don't know what's
to happen.
Hugh Wil (40:00):
A lot of conversations right now about sovereign AI
So, you know, a lot ofcountries thinking, hey, we should
our own AI capabilitythat's totally walled within our
so that we control thisinfrastructure, we control what it
do, we can make sure thatthe laws that we have are applied,
we can make sure that ourdata stays in our country.
So certainly lots ofdiscussions about that right now.
(40:22):
And I think governmentsand institutions are getting more
more uncomfortable withthe idea that their data could be
anywhere in the world andyou know, stored even in the US.
Hann (40:31):
Yeah, I think I saw France the other day were the latest
to announce like a for themwhat was a significant investment
in AI.
Although I think thecontext in which it came up, someone
saying, yeah, it sort ofpales in comparison to what these
US tech companies are making.
And I guess that's where itbecomes difficult to establish
If you're a country thatwants to play in this pond, right,
I'm sure everyone does forreasons we've mentioned, if you're
(40:54):
not well established withthe capability or the investment in
it, you're startingbasically on the back foot.
Hugh (41:01):
You need the intellectual capital, you need the the actual
to really invest, youneed to go and find people and get
to stay, have the rightregulatory environments, all these
of things, the datacenter infrastructure.
So, you know, very, veryhard to replicate something like
Coast tech, you know,anywhere in the world.
And and indeed, you know,if you're going back to the
Seven means a good way toround out the show, the market cap
(41:23):
of each one of those ismore than the GDP of, you know, most
nations on Earth.
Hannah (41:28):
Yeah, it's pretty crazy.
Like they make up 30% ofthe SP 500, which on a global scale
you know, outsized again.
And anyone that's investingin like an index fund, you know, if
you've chucking yourpension or some of your savings into
tracker or something likethat, you're going to be investing
these companies.
So you kind of can't getaway from them.
Hugh Williams (41:46):
Absolutely.
Hannah Clayton (41:47):
All right, guys.
Well, apologies if any ofyou listened to that on your morning
commute and you'rewondering if you've made the right
Hugh Williams (41:54):
But thank you for listening.
And we love, we love allof our listeners out there who
to the Tech Overflow podcast.
You can find us on socials.
We are on LinkedIn, we'reon X, we're on Instagram, we're also
on TikTok.
Hannah (42:08):
We are on TikTok, guys.
Time to get an account.
Hugh Willia (42:10):
And YouTube Shorts.
So you can find us allover the web and if you want the old
school version,techoverflowpodcast.com.
And we're of courseavailable on all of the podcast
Han (42:19):
And if you like what you've heard today, do recommend the
to your friends and family.
Give us a like, subscribe,and leave us a review.
We love to hear from you.
Hugh Williams (42:28):
Yeah, absolutely.
Go to Apple Podcasts,write a review if you really want to
thanks for the podcast.
Han (42:33):
And guys, you heard it here first.
It's time to retrain as adata scientist.
So I better go and lookthat up now.
Hugh Williams (42:37):
Yeah, awesome.
I'll help you, Hannah.
Hannah Clayton-Langton (42:39):
Awesome.
Let's go and do that.
Thanks, guys.
See you next week.
Hugh Williams (42:41):
See you soon.
Bye.
Why does it matter if the person gets the code that they need
at a price that they thinkis fair?
Hugh Wil (00:07):
You can vibe code up a demo to demonstrate any feature or
product idea that you have.
So turn up with somethingthat works.
Hannah (00:13):
For someone who doesn't work in that industry, that stuff
really shiny and sexy andappealing.
Hugh Will (00:20):
world, and welcome to the Tech Overflow podcast.
My name is Hugh Williams.
Hannah C (00:24):
And I'm Hannah Clayton Langton, and we are the podcast that
down technology for curiouspeople.
Hugh Willia (00:30):
How do I do Hannah?
Hannah Clayton-L (00:32):
You did great.
Uh that felt weird.
Guys, we decided to do someA-B testing and mix up who does the
intro.
So if we get much morelistens and engagement on this
we'll make that a permanentswap.
Hugh Willi (00:42):
Or we'll try another experiment, one word each, or I
I'm not easier.
Hann (00:46):
And guys, this is a signal that I am now an official tech
and I will be leading thisepisode as our resident technical
Hugh Williams (00:53):
Yeah.
So our topic today, uhHannah, is transistors, the
of the transistor.
So um tell us about transistors.
When were they first invented?
Hannah Clayton-L (01:00):
Okay, correct.
I've been outed pretty quickly.
Okay, we will do an episodeon transistors later on, let's say
season seven.
be by then.
So, no, we're not talkingabout transistors today, guys.
We are talking about big tech.
Hugh Williams (01:15):
Yeah.
Obviously, I spent a lotof my career working in big tech.
So I'm really uh excited,Hannah, to talk about all the
the West Coast tech,maybe a little bit about Taiwan,
and really give our smartlisteners out there some insight
how these companies operate.
Hannah C (01:27):
Guys, my goal for this episode is to get you to spill as
secrets as I can aboutSilicon Valley.
So wish me luck in Endeavour.
Um, I reckon we getstraight into it because I really
mine you for gossip.
no episode about big techwould be complete without a
to the Magnificent Seven.
So those are the sevenbiggest tech companies in the US,
(01:48):
listed, I guess, biggesttech companies in the US.
And I'm sure you can dothis, but I'm gonna ask you to name
seven to kick us off.
Hugh William (01:55):
Yeah, so Apple, of course, uh, Microsoft, Amazon,
which owns Google,amongst other things, also owns
an episode on in season one.
Uh Meta, which of courseowns Facebook, Instagram, and
NVIDIA, which make uh theGPUs that uh we've talked a lot
(02:15):
in season one, and Tesla.
Hannah (02:18):
Ding ding ding, got them all right.
So they make up, I think,about 30% of the SP 500, which is
US's largest 500 publiclylisted companies.
There'll be a few on therethat listeners will know that we
mentioned just becausethey're not publicly listed.
So like OpenAI, Anthropic,right?
They're not.
Hugh Williams (02:34):
Yeah, SpaceX.
I mean, there's there'squite a few large companies that,
know, may publicly listin sometime in the near future.
Uh, but they're the majorones that are publicly listed.
And Magnificent Seven's anew moniker, if you like.
It hasn't been around that long.
We used to use themoniker Fang, which had two A's in
I think is Facebook,Amazon, Apple, Netflix, and Google.
They were the cool kidson the block probably five years
(02:56):
but now we talk about theMagnificent Seven instead.
Hannah Clayton-Lan (02:59):
Interesting.
So Netflix dropped out.
And I'm surprised that youdidn't mention Microsoft.
I would have thought theywould have been in there.
Hugh Wil (03:04):
They were a really big comeback, really, under under Satya
leadership.
And the Azure business isa very, very big business.
They're cloud business.
And we talked about clouda little bit in season one,
in our in our wrap episode.
And also, you know,they've been major investors in AI.
They're a largeshareholder of open AI, famously.
And so, you know, they'vereally made a comeback, and so
(03:24):
part of the Magnificent Seven.
Hannah Cla (03:26):
Interesting that you mentioned cloud because I did a
of research on the revenuestreams of the Magnificent Seven.
And some of it's quiteintuitive and some of it's a bit
So yeah, Microsoft, a lotof the profit driven by cloud.
Speaking of cloud, I wassurprised to learn that I think it's
something like 75 or 80% ofAmazon's revenue is still coming
their retail business, buttheir profit margin is mainly driven
(03:47):
by AWS.
So things like cloud, Iassume, are a much better play than
know, selling bits and bobsonline to consumers, but then the
retail business becomes thevolume play.
Hugh Willi (03:58):
Yeah, yeah, exactly.
I mean, low margin, highvolume business, the retail
And then because of thescale of AWS and the ability that,
know, you can really getscale from that infrastructure, the
margins are much, much,much higher.
And they really are thecompany that largely sits under most
other companies thatyou're using on on your smartphone
on the web.
Han (04:18):
And Google have got a cloud business, but they're, I think
like 90% of the revenuecomes from ads, which I think we're
to talk a little bit moreabout Google in an upcoming episode,
but that's not a hugesurprise given the power that sits
their data collection.
Hugh W (04:31):
One of the best business models, Hannah, of all time is
the Google Ads business.
Han (04:35):
Well, we actually should do a whole episode on that.
Hugh Williams (04:38):
We will.
Hanna (04:38):
And then I was interested to learn that Tesla actually make 10
to 15% of their revenue onlike environmental credits, which I
assume is just sort of away for less environmentally
companies to offset some oftheir emissions, like the sort of
old school car productioncompanies or something like that.
Hugh Willia (04:54):
Yeah, yeah, I think that's the case.
And look, historically,Tesla certainly benefited a lot from
you know governmentcredits, government programs,
kinds of things.
So it's a you know it's acomplex company.
Obviously, it's buildingin its end product something that's
incrediblyenvironmentally friendly, an
that can be used to powerall kinds of things.
Long conversation aboutwhether the actual production of the
car and the batteries isenvironmentally friendly.
(05:15):
I think that's a that's aseparate conversation, but certainly
they've benefited a lotfrom you know regulatory and and
support.
Obviously, a lot ofthat's been wound back by the Trump
Hannah Cl (05:25):
Uh, we won't get into politics overflow on this podcast.
Find us in the pub for thatconversation.
Microsoft's an interestingone because I don't know how widely
known their investment inopen AI is sort of outside of the
world.
And also, I mean, less sexythan ChatGPT, but um, they own
which I only learned when Iwas prepping for this episode.
Hugh Williams (05:43):
Yeah, yeah.
Looking, you know, theyown I think 49% of OpenAI.
And yeah, I think they'veowned LinkedIn since probably around
2016, Hannah.
Hannah Clayton (05:52):
That is correct.
And they bought it.
Can you guess how much for?
Hugh Willia (05:55):
Oh, look, my mind's saying 9 billion, but I bet it's a
bit more than that.
Han (06:00):
It was $26 billion, which I I mean anyway.
Off by an order ofmagnitude, that's me.
Well, I would have guessedmuch lower.
Um, anyway, so yeah, sosome interesting sort of makeup of
revenue of those companies.
You can certainly, youknow, dig a bit deeper.
We won't get into that anyfurther for now.
And then, of course, theMagnificent 7, those are all by
US-based.
We do see a huge amount ofbig tech innovation come from the
(06:23):
coast of the US, and we cantalk about why that is maybe a bit
later on.
But maybe a nod to Taiwanfor the semiconductor production.
So TSMC, I believe thatcompany's called.
And then we've got ByteDance who own TikTok and are
based in China.
More on them later.
And I think it'd probablyfair to call out Samsung in Korea.
Do you think we've missed any?
Hugh Willi (06:43):
Yeah, we've probably missed quite a few.
I mean, there's a lot oflarge Chinese companies, Ali Baba,
Tencent, you know,there's a long, long list.
We should do an episodeon Chinese tech.
And then, you know,there's there's companies that
large as those that arebased outside the US that are pretty
famous.
So, you know, an examplemight be uh Spotify.
Yeah.
Hannah Clayton (06:59):
Yeah, that's it.
I had didn't realise for along time that they were Swedish.
I think that's super cool.
Hugh (07:03):
But on TSMC, I mean, they are the company that really produces
the silicon for most ofthe Magnificent Seven.
So, you know, the supplychain from Taiwan is very, very
to US tech.
And again, that'sprobably a topic we could dig into
season.
Hannah Clayto (07:16):
Yeah, definitely.
The the geopolitics of itsuddenly become quite scary.
Um, but that's not whatwe're gonna focus on today.
I want to take us back tothe west coast of the US because
where Hugh you spent a lotof your career.
But like, why do you thinkwe got these hubs of huge tech
coming out of it?
It's Seattle as well asCalifornia.
I was gonna say maybe itwas the weather, but I think the
(07:36):
in Seattle's a bit crap.
But um, yeah, like why doyou think those hubs ended up there?
Is it just because it'swhere Bill Gates was from?
Hugh Wi (07:42):
You know, the Microsoft story is really the Seattle story,
it's probably not superwell known.
You can read about it inum Bill Gates' latest book, which is
called Source Code.
You know, Microsoft'searly years were really in
New Mexico, and then Billand Paul Allen decided to actually
move the headquartersback to Seattle, which is where they
both were from.
(08:02):
So it was almost anAlbuquerque, New Mexico business
for a while there itbecame the center of tech in the US.
Hann (08:09):
Well, uh Amazon is Seattle too, is that right?
Hugh Willia (08:12):
Yeah, that's right.
That's right.
And and you know, I thinkit's a beneficiary of the fact that
Microsoft was there andreally turned Seattle into a into a
tech destination.
Hannah C (08:19):
Because once you get a critical mass of good engineers, I
is mostly what are indemand, then you want to co-locate
can try and steal that goodtalent, right?
Hugh Williams (08:28):
Yeah, absolutely.
And you know, if you workin any of these companies, you'll
find that there's asignificant number of people who've
there for their wholelife, but most of the other people
come from one of theseother large tech companies.
So there's lots of peopleat Google who are from Microsoft,
lots of people atMicrosoft who are from Google,
know, that these folksjust move around.
Han (08:44):
Okay, so Seattle, that sort of makes sense.
And then we travel down thecoast to California.
Why do you think that built up?
Hugh W (08:51):
Yeah, we fly over Oregon and get to and get to California.
Hann (08:55):
Funny that, not a huge hub of tech, globally, Oregon.
Hugh Williams (08:58):
Yeah.
And if you've ever livedin California or have been to
you know that it's reallytwo states.
So they call it NorthernCalifornia or NORCAL and Southern
or SoCal.
And you know, they'rereally different places.
Um, arguably it could betwo states, you kind of rule a line
across the middle.
And LA is perhaps famousfor the film industry and all those
kinds of things, themusic industry.
It's a very, verydifferent culture to the culture you
(09:19):
find in the Bay Area,which is the area just below San
which is, you know,famously known, I suppose, outside
Bay Area as the Silicon Valley.
So you'll find the peopleinside the so-called Silicon Valley
just call it the BayArea, and uh the rest of the world
it the Silicon Valley.
But I think the historyreally is that we had Stanford
obviously one of the youknow, world-class university.
(09:41):
It's always been very,very good at computer science, and
it's always produced topshelf graduates.
Larry and and Sergei, whofounded Google, are, you know,
folks.
So lots of lots of famouspeople have come out of Stanford and
gone on to foundcompanies, but it's historically
very strong school forcomputer science.
Uh, defense was alsobased sort of in that region.
(10:01):
So there's lots ofdefense industry there.
Climate's great.
Back in the day, therewas plenty of land, lots of fruit
And so, you know, you puttogether the universities with the
defense investment, witha wonderful climate, plenty of land.
And the end result is itbecame really the center of tech in
the world.
Han (10:17):
Okay, so it literally comes down to accessibility of smart
I know from having workedwith robotics through my current job
that there's like a big hubaround Boston for similar reasons.
You've got like Harvard andTufts and Boston College and some
of the really good unis.
So it sort of makes sensethat that's where the employers
to.
And what years were you in?
You were like early SiliconValley, right?
(10:38):
Is that fair?
Hugh W (10:39):
No, no, no, no, not fair at all.
No, no, Silicon, no, I'mnot fairly.
Maybe we can take that out.
Hannah Clayton- (10:43):
We'll let's go.
Hugh (10:44):
No, I'm definitely the uh senior member of the team here, but
I'm not quite that senior.
Hannah (10:48):
Okay, so what years were you in Silicon Valley, aka the Bay
Hugh Wi (10:52):
Yeah, so um we moved to the Bay Area in 2009, and I was
until 2017.
So just a small part ofthe history of uh of the so-called
Valley.
Hanna (11:04):
But that was a really big so maybe just since you've mentioned
timelines, just to placesort of what was going on in the
of tech, because I I foundthis quite interesting as it came
of my research.
if we think about thehistory of what tech meant, right?
Because back in the day, soin the 1980s, let's say it was all
about personal computing,the internet wasn't really a thing
or not something that theeveryday person was using.
(11:24):
And then the 90s was allabout the internet boom.
I I literally remember thefirst time we crowded around my
desk, um, actually also inthe US, and dialed up to the
with that very distinctivesound.
If you know, you know.
2000 to 2010 was all aboutsocial media.
And then we moved into appsand the app economy and app for
And now when we talk abouttech, we're really talking about
(11:46):
cloud, generative AI,platform tech.
So through that timeperiod, tech as an industry meant
things depending on whatwas going on.
So you were there duringthe sort of social media, mobile,
app economy phase of things.
Hugh Williams (11:59):
Yeah, that's it.
And probably the birth ofcloud computing, really, as it
something that becamecommoditized and available to other
around the world.
And we've spoken a littlebit about it in earlier episodes,
Hannah, where we'vetalked about sort of how Microsoft
its own cloud for its ownpurposes, and in its case for Bing,
and Google did the samefor its Google search, and Amazon
the same for its store.
And then obviously theywent on and sold those things to
(12:20):
people, and that's reallynow what powers the internet.
But certainly I was therefor the the mobile revolution, maybe
some part of the webrevolution and definitely the cloud
Hannah (12:29):
And does that change the type of people like computer
that you need?
Do you need differentskills or is it all covered in a
science degree?
Hugh Williams (12:38):
Yeah, really good question, Hannah.
I mean, if you zoom out alot, right, I think at the beginning
of computing, it was allabout hardware.
And you know, softwarewas a small part.
You know, certainly overtime, what's happened is hardware's
become, you know, it'sstill interesting, but the the real
has gone into software.
And look, I think a goodcomputer science degree gives you
problem solving skillsthat you can use to adapt to any new
(12:58):
technology.
So that's one of thewonderful things about doing
is you learn how to be aproblem solver.
Um, you learn how todecompose a problem down into
and solve those problems.
You know, we talked alittle bit about that way back in
first episode of season one.
And so, you know, atleast on paper, I guess uh somebody
a computer science degreeought to be able to adapt to changes
in technology.
But I say on a morepractical level, the kinds of skills
(13:20):
are valuable has haschanged quite a lot.
I mean, there wascertainly a phase when maybe we'll
back to the start of theweb where being a web developer,
could build web pages andmake those interactive was kind of
the the hot skill.
And then being an appdeveloper became a very valuable
You know, if you couldwrite iPhone apps or Android apps,
was a valuable skill.
Again, that's a littlebit more commoditized now,
(13:41):
with things like ClaudeCode, where you can generate large
of code.
And these are often verysimple apps to build.
Then we sort of wentthrough a phase of where the sort of
infrastructure people whocould stitch together AWS and Azure
and Google's cloud toreally build scalable systems were
valuable.
Those folks are stillquite valuable.
But what I think whatwe've really seen over the last few
(14:02):
is the emergence of thedata scientists.
So these are people whocan use data to create products, and
they're often people whoare very skilled at machine learning
and AI.
And obviously that is theuh topic de jour, if you like, of
uh of computer science atthe moment.
Hannah (14:16):
Yeah, I wonder how that will develop.
If you think about it, likeif someone works for 40, 45 years
in their career, that'sspanning a lot of different eras
tech.
And so I guess you eitherhave to be, you know, super
to the thing that you do,even if it becomes out of fashion
sort of the main, or youhave to be pretty agile, right?
And like, would people retrain?
Hugh Williams (14:35):
Yeah, absolutely.
I mean, the programminglanguages that I learned at
are largely not used anymore.
So certainly, you know,through my career, I've had to adapt
to different programminglanguages, different ways of
software, um, you know,pretty much everything's changed.
But again, if you've ifyou've got those fundamental
science characteristics,you should ought to be able to, you
know, adapt to change over time.
And that's what I alwayssay to people when we're hiring, we
should be hiring forfundamental sort of competencies, if
(14:58):
like.
Is a person a goodproblem solver?
Are they smart?
Are they driven?
They got a computerscience degree, not hiring for
who has a specific skillbecause you know, skills come and go
very, very quickly in uhin the tech industry.
Hannah Cla (15:10):
these highly skilled problem solvers that you're drawing
top universities on thewest coast of the US, they're
paid a lot of money, right?
Like if you're working intech, if you're an engineer, there's
sort of two paths you cango down.
Maybe this is my outsideinterview, and you can go for like
lifestyle, work remotelyfrom a beach, get paid pretty good
considering that you'resort of in control and have the
or you can go after like acold, hard cash and land a job in
(15:32):
Silicon Valley and make alot of money.
Hugh Williams (15:35):
Yeah, that's it.
I mean, I think there'slots of things you can do in tech,
certainly if you want tobe amongst the best of the best and
be paid the largestamount possible, then you know,
the Silicon Valley is theis the way to go.
Hannah C (15:45):
And it's not just like monetary compensation, like Netflix
have unlimited holiday.
And I know that Google umsupport female employees freezing
eggs, which is actually areally awesome benefit that I have a
couple of friends that havetaken them up on it.
Free food.
I remember I once worked atYouTube early on in my career.
I wasn't working forYouTube, but we did an event with
(16:05):
and it was like going intothe Google offices and seeing that
they do actually have thenap pods and the massage room and
free food that you can justtake from the canteen.
Like for someone whodoesn't work in that industry, that
was like really shiny andsexy and appealing.
Hugh W (16:19):
Yeah, and look at all of it's true.
Certainly when I workedat Google, I worked at the Mountain
View campus, so the maincampus of Google.
I think it probably hadaround 50 different restaurants.
Wow.
Um, not all of them openfor all meals, but you know, they
breakfast, lunch, and dinner.
Pretty much select anycuisine you want any time of the
So I used to hit the gym.
They they probably had, Idon't know, more than 25 gyms.
(16:40):
They'd sort of done afaux athletics track with the
grass and things.
They had big tractortires there that you could flip
that you could hit thingswith, put out sort of the cones and
have running drills, youknow, a bit sort of boot camp kind
of style.
I used to enjoy that.
And when I finished theclass, go and have a shower and uh
the free towels, ofcourse, uh that come with the uh the
(17:01):
And then I'd head over toone of the smoothie stations and uh
get them to make me up areally, really good protein smoothie
with all the additivesthat that I wanted.
Usually for lunch, youknow, I'd pick one of the healthier
quite like one of theJapanese restaurants they had, used
go there.
And then I didn't usuallystick around for dinner, I'd jump
on my bike and pedal home.
But certainly, you know,the culture is designed to get you
in there early so you geta free breakfast, feed you at
(17:23):
and then get you to staylate, have dinner, stick around,
do a little bit more work.
So, you know, the cultureis designed to get you to work hard
by providing you with theuh the perks to help you do that.
Hannah (17:34):
Okay, I would be such a sucker to that culture.
Like, if you give me a goodgym on, I'll be there on the
Like that is very smart.
I'll take us for a minute alittle bit further away
But my brother worked forbite dance in China for a while.
Oh, wow.
And um I asked him for someof the sort of insights into how it
was working there becausethey worked him pretty hard.
(17:54):
I think he was there fortwo or three years.
And he said similar things.
So it was like core workinghours were 10 till seven, I think.
But free breakfast was fromnine and free dinner was from seven.
Hugh Wil (18:04):
So it's like nine nine six is they call it him.
Hannah Cla (18:06):
Yes, yes, that's it.
Sorry, I should have openedwith that.
Beijing tech is the ninenine six, so you work nine till nine
six days a week.
Yeah, yeah.
Um, actually, sorry, juston the six days a week thing.
He said that when he movedto work at Byte Dance, they were
changing from working sixday weeks every week to working a
week and then a five-dayweek the following week.
And they did somethingreally cheeky where they were like,
(18:26):
we were actually paiddouble for Saturday, so you're gonna
you know, more thanone-sixth of your pay cut on the
you are only doing five days.
And he said that somepeople really didn't like that
just actually wanted towork super hard.
He said there was a ton ofyoung people who worked there
until they burnt out.
And then he said they gotsubsidized rent, but only as long as
it was within 20 minutes ofthe office or something.
(18:47):
So he didn't go for that one.
He actually said that itinflated the rents in the local area
and so it wasn't reallyworth it.
But it was that wholedesign of like, it sounds really
someone like me who wants aconvenient gym will be like
bought in.
But it's basically designedto keep you on site and working for
as much of the day aspossible, right?
Hugh Williams (19:06):
Yeah, absolutely.
And going back to theGoogle story, I mean, they had
much had every serviceyou could possibly think of, and
were free.
So they had like a bikerepair store.
You could, you could, youknow, if you had an issue with your
bike, you could take yourbike there and they'd repair it for
you while you're at work,laundry, all these kinds of things.
Dry cleaning?
Uh yeah, yeah, certainlydry cleaning.
Um, I can't rememberwhether it was free or not, but
they had, you know, drycleaning where they'd come and pick
(19:26):
it up and drop it back.
Um, and you know,probably subsidised at the very,
But certainly, you know,every service you needed to
live your life wasavailable on the campus again to get
to stay on the campus.
The only thing theydidn't have was beds.
So you couldn't actuallysleep overnight there.
Hannah Clayton-Langton (19:41):
Okay.
And did they have likechild care and or pet care?
Like sometimes you see thatlike doggy day care is uh in London,
like dog-friendly officesis a big pull for certain people.
Hugh Wi (19:50):
Yeah, so certainly, you know, all of those things are
of.
So if you sort of zoomout a little bit and think about
of the broader benefitspackage for somebody working at
I mean, certainly theywant you and your family to be
So the health benefitsare fantastic, you know, and a lot
things like, you know,looking after your pet, childcare,
kinds of things are allpretty heavily supported.
And obviously, becauseit's one of the most wealthy places
the world, there's anenormous number of workers who come
(20:13):
that area to providethose kinds of services.
So it's definitely thecenter of the universe when it comes
to sort of being lookedafter, both within the company and
you know, broadly as afamily that lives there.
Hanna (20:24):
you get the same perks if you're like a receptionist or like
a finance manager?
Do you get the same perksas the engineers?
Hugh Williams (20:30):
Yes and no.
I think um there'sdifferent colored badges.
And so you know, theemployees get a get a badge that's
colour and thecontractors get a badge that's a
and you you get virtuallyno perks if you are a contractor.
And you would find, forexample, that pretty much all the
are going to becontractors, contracted through a
And so they're wonderfulpeople, they work hard, you know,
share the ethos, but theydon't get any of the benefits really
(20:53):
that the employees get.
Hannah (20:55):
I had heard that Google, because Google got big offices in
that they mainly hirepeople at least to start on
And I had understood thatthat was because it's easier to sort
of end the relationshipwith the employee if it's not gone
and then they can just makeyou permanent if it does.
But I hadn't consideredthat's probably also a benefits
Hugh Wi (21:12):
Yeah, look, it probably is.
And and look, you know,California has um what we call
deployment.
And that basically meansthat they can get rid of you or you
can quit and you can dothat fairly quickly.
Obviously, places likethe UK and Australia, it's a lot
difficult to get rid ofemployees.
Uh, and you know,sometimes, particularly in the UK,
an employee to leave.
So, you know, some of thesenior executives in UK tech have
(21:34):
to give crazy noticeperiods that, you know, like 12
right?
Yeah, just completelyblow my mind.
Where, you know, the Ithink the longest I've probably ever
seen in the US West Coastis probably three weeks.
Hannah Clayton-Langton (21:43):
Oh, wow.
Okay.
Wow.
Even UK, the shortestyou'll get is one month.
And back to these differentcolored badges because I'm kind of
shook by this.
Okay, so does that createlike a two tier Diddy engineers who
I presume are the people onthe permanent contracts, like, does
that make them sort of likefirst class?
Citizens and then the otherones like maybe less so.
Hugh Williams (22:03):
Yeah, I think uncomfortably so.
I think uncomfortably so.
Hann (22:06):
Okay, so we've kind of got a two-tier system.
And so would your likein-house lawyers be contractors or
they be full-time?
Hugh Wi (22:12):
No, they'd be they'd be full-time.
They though things likepatent lawyers, you know, the folks
who um who write up thepatent applications and things would
be external legal firms.
So when it comes to sortof some of the specialist legal
you know, they use thesewell-known external firms, but
all the professionalstaff who are paid well are
Hannah Clayton-Langton (22:31):
Okay.
And it's obviously pairedwith like caning the working hours,
right?
Like I imagine if you are avery highly paid engineer at one of
the big tech firms, theexpectations on you are pretty high.
And what you've said aboutour well employment means that that
can all be ended prettyquickly by an employer if they don't
like they're getting thevalue from you.
Hugh (22:48):
Yeah, these are these are performance cultures for sure.
Um, and I think most ofthese companies are by any measure
class at setting, youknow, KPIs or performance
of the people that work there.
They by and large havepretty strong review cultures.
So, you know, you sitdown and you talk about your
They have very strongbonus cultures where they'll pay
(23:09):
or not pay bonuses basedon people's performance.
And so certainly theseare cultures where, you know, people
are encouraged to workhard towards very specific goals.
And, you know, if youdon't hit those goals, it might not
you bonus.
And uh, you know,probably not far behind that is you
be shown the door.
Hannah (23:24):
Okay, and talking about bonuses, like what percent of
base would they then bestriving for as their bonus?
Hugh Willia (23:31):
Yeah, so look at it depends on level.
If you're a recentgraduate and you you join these
you know, your bonustarget might be anything from say
to 10 or 0 to 20% of yourof your base salary, which, you
sounds pretty good tomost people.
Once you get up into thesenior executive ranks, you know,
probably in the range ofzero to a hundred percent, maybe
to two hundred percent ofyour base salary.
(23:51):
And so, you know, youstart to think of the bonus as the
part of your cashcompensation, though a very
of your cash compensationbecause it doesn't just depend on
your personalperformance, but it also depends on
performance.
And obviously, you'resomebody who might have some
on that, but you don'tcompletely control that.
Hannah C (24:08):
Okay, but that kind of makes sense at an executive level
the bonus targets would bethat high, right?
And that's not specific totech necessarily.
But what about like asenior engineer?
Like they need to beinvested in what they're building,
way that we do that is bygiving them shares.
That's another part of compwe've not really mentioned shares
or bonus.
Like it do you find thatthat that mid-range gets pretty well
(24:29):
looked after in that senseas well to get them bought into what
they're doing?
Hugh Williams (24:33):
Yeah, absolutely.
And you know, salaries goup very, very quickly in in US tech
from uh junior salarieswhere, you know, you're an engineer
and you're perhapsstarting your first role at a place
a TikTok, you might begetting a base salary of, say,
US, which sounds good byany international standards.
And then once you're inthe door, if you're somebody who
hits those targets, keepsgetting bonuses, I mean, you could
(24:54):
certainly over a periodof years be, you know, getting paid
up around a milliondollars to be an individual
you know, somebody who'ssitting there writing software every
day.
Hannah (25:04):
Yeah, guys, I picked the wrong career path.
Well, who knows wherepodcasting will take me, but
feel I went to an all-girlsschool and it would we were sort of
pushed if you were intoscience that you'd do medicine.
And if not, don't botherwith science.
And there was a big gapwhen it came to things like
computer engineering, maybeit would have been called.
And I think my schoolmissed a trick.
Hugh William (25:22):
Yeah, I think most schools miss a trick, Hannah.
I mean, we should talkabout our charity code for schools
some point and what we'retrying to do here in Australia to
change that trajectory,particularly for girls.
But uh, yeah, lots ofschools don't understand how
tech is and uh how muchyou know one engineer can change the
world.
I mean, you really canwork on a product that affects
of people, and andthere's really very few industries
(25:43):
that's the case.
And so, you know, it's ait's a pretty interesting career to
get into, and because ofthat, leverage that impact, you can
you can make lots ofmoney for yourself too.
Hann (25:50):
Well, think about Jonathan Medine, right?
We had him in episodefive-ish of season one, and he just
one night coding, inventedthe swipe right, and that's
changed the face of moderndating.
So it's definitely not anover-exaggeration to talk about that
impact.
And I have many a timeconsidered retraining.
Think about it all the timethis episode too.
Giving me some more pausefor thought on that as well.
(26:12):
And I think just whilewe're on the topic of the sort of
split, which we we don'tneed to get into in too much detail.
But one thing I have heardis that the female engineers that do
exist, because they're sortof famously, um, it's more of a male
skewed industry, but theycan demand some pretty big salaries
because the big companieshave quotas or they want more female
engineering presence.
And so if you're a strongfemale engineer, you can sort of go
(26:33):
in and demand your priceand they'll be really fighting for
Hugh Wi (26:36):
You know, certainly all the organizations that I've worked
and this concludes inrecent times, they're probably at
90% male and you know,maybe around 10% female, and that's
know, clearly that's notrepresentative of, you know, how the
world is.
And so I think morediversity would be fabulous in tech,
certainly uh, you know,you have to encourage women to get
tech, and then uh thosewomen who get into tech and who are
great at tech, you know,everybody wants them quite
Hannah Clayt (26:59):
Yeah, fair enough.
Okay, girl power.
So back to shares.
Does that not dilute theshareholding of a company if they're
just giving loads of freeshares out to all their employees,
is it not quite as simpleas that?
Hugh Wi (27:12):
It does create dilution in the in the shares, but probably
to a significant extentunless you're really talking about
who are getting, youknow, big parts of of the company.
But it's a very, veryimportant part of the compensation
anybody in the US.
And there's really threecomponents to the comp in West Coast
Tech, Hannah.
And, you know, even youngengineers are pretty savvy about
So, you know, first ofall, there's your base salary.
(27:33):
That's the cash you getevery bi-weekly, as they'd say, in
US.
The bonus.
We talked a little bitabout that.
Young engineers will beinterested in what's the target
and what do I have to doto get it?
Do you ever pay more thanthe target bonus?
Those kinds of things.
And then there's theRSUs, the restricted stock units,
if you like, of the company.
And most engineers willbuild a little spreadsheet and
say, good, better, best.
(27:54):
They'll put in somepossible stock prices, share prices
might be reached.
They'll build out thevesting schedule of those shares and
figure out how much theycould be potentially worth.
And that's certainly areally big factor in the
process of engineers isyou know, do they believe in the
prospects of the companyand will they make a lot of money
of their shares?
Hanna (28:11):
And and RSUs, or at least my experience with them, was that
vest over a certain numberof years.
So it's like, oh, great,we've given you this huge slug of
but you've got to stay herefour more years to get them all.
Hugh Williams (28:21):
Yeah, absolutely.
It's a little bit morecomplex than that, too.
So let's take a fictionalexample.
So imagine you uh youjoin a company and they give you
RSUs as a as an awardwhen you join.
So they say, here you go,you can have 10,000 shares.
What'll typically happenis that they'll vest out over a
period.
There's often sort of acliff before you get the first part.
So it might be say, hey,you have to be here six months
(28:43):
you're gonna get any, andthen you'll get six months' worth.
And then maybe they'llvest every month after that until
reach the four years andyou've got all 10,000.
But then what'll happenevery year usually in these
companies is they'll giveyou another share award.
So they'll say, Hey,Hannah, you've been fantastic.
We're gonna give you14,000 shares this year.
And again, they'll vestout over four years.
(29:05):
And so now you've got twolots of vesting going on.
You've got the awards yougot on the first day you joined, and
then you've got someaward that you got on the 365th day,
anniversary of a joining.
And then guess whathappens next year?
We'll give you another pile.
And so you've got allthese things vesting on different
different numbers of shares.
Han (29:22):
Different share price, like share price fluctuations, right?
Because if there was a youalways want shares on a down year if
you believe in the company,because that's where you're going
to make most of your money.
Hugh Williams (29:31):
Absolutely.
So, yeah, often whatthey'll do is, as you say, is
well, we want to giveHannah a a certain dollar value
of shares.
And so they'll dividethat by the current share price to
the whether it's 10,000shares or 12,000 shares.
And so you're factoringin, you know, the future prospects
the company, the numberof shares you might get, but you've
got this very complexequation going on in your head.
You know, they call itthe golden handcuffs, because
(29:52):
you leave, you lose allof those shares that are yet to
Han (29:56):
And arguably the best share play is to go for like a super early
phase startup that you justbelieve in, right?
Like I had a an oldcolleague once who he, that was his
thing.
Like we were on a gradscheme together, and then he was
My next move, I'm gonna goearly phase, be the single digit
number, and then I'm gonnamake a ton of money once that goes
big.
But basically, if you'rereally good at spotting, you know,
(30:17):
next unicorn, that could,you know, make you could make your
by the time you're 30.
But I imagine that there'smore people who don't than do who
that approach.
Hugh Williams (30:25):
Absolutely.
And different, you know,it's just about different levels of
risk appetite, right?
So there's certain peoplewho are happy to go for that hero
or zero company where,you know, it's either gonna be worth
nothing at all, or maybeyou'll make a fortune.
And there's a certain setof people, and I probably fall into
that category of who arehappy to work for big tech.
You get a generous basesalary, you get a predictable bonus,
you get some RSUs orstock units that you know are going
(30:46):
be worth, you know,roughly in this kind of range,
something special happened.
And something special hashappened over these last few years
as share prices of theMagnificent 7 have run up
Hann (30:56):
All of this must amount to quite a distinct culture in Silicon
Valley, and I'm sure inother sort of tech hubs that exist
in the world.
But like it sounds likethese people might be a little bit,
don't know, like maybearrogant, which isn't a word I would
associate with likeengineering, you know, sort of
but it's kind of crazy thatthey would be making all this money
(31:18):
and be in this bubble whereit's normal to get all this sort of
compensation and it's quitedistinct from the real world that
most of us live in.
Hugh Williams (31:26):
Yeah, yeah, yeah.
And I recommend the theTV show Silicon Valley if you kind
want to see a caricatureversion of this.
It's uh it's pretty funny.
But yeah, certainly, Imean, you know, when I was a kid, I
just kind of some nerdy geek.
And then when you move toSilicon Valley, you know, you're
of a population of peoplelike you, and you know, you feel a
little bit like a rockstar given the money that you're
paid and where you liveand all the things that you can do
(31:48):
that kind of money.
Hannah (31:49):
Okay, so this episode is now sounding like a pretty strong
for people becomingsoftware engineers.
about the disruption ofsomething like Claude Code?
Like suddenly you've got amuch cheaper way of generating that
output.
Even if you don't removeall of the engineers, you remove 50%
of them, 70% of them, thenit's much cheaper to use that.
And then your big ticketsalaries and compensation and
(32:10):
packages might just disappear.
Hugh Willi (32:13):
Look, I think that's partly true and partly not true.
And I think it's a littlebit unclear exactly where this is
all headed.
But if we if we wound theclock back, say six months, I had
the very strong view thatyou know, things like clawed code
were kind of generating,you know, code that could be used in
prototypes to demonstratea concept, but it really wasn't
beyond that.
So everybody's quoteunquote vibe coding up these apps
(32:35):
saying, hey, softwareengineering isn't too hard.
And I'm like, try gettingthat running at scale where you have
99,000 queries a second.
Hannah Clayton-Langton (32:42):
Okay.
And then six months ago,you you're thinking, eh, like good
prototyping.
Yeah.
Hugh W (32:47):
And good for prototyping in a super cool way.
Like we spoke aboutproduct management back in season
think it was our second episode.
I think even six monthsago, you know, my advice to young
managers is don't turn upwith slides, don't turn up with
don't turn up withPowerPoints, turn up with a demo.
Like you can, even ifyou're not a really proficient
you can vibe code up ademo to demonstrate any feature or
(33:08):
product idea that you have.
So turn up with somethingthat works.
Hannah Clayton-Lan (33:11):
I love that.
That's really cool.
Hugh Williams (33:12):
Yeah, yeah.
And I think we shouldexpect that of product managers
It's like go build it.
If you think it's real,build it and show me the demo.
There's even a course nowuh at CMU, Carnegie Mellon
called Vibe Founding.
Hannah (33:23):
What and what does that mean?
Hugh (33:24):
So the idea is, you know, you have a startup idea, and within
a week or two of runningin this class, you actually come out
with a product that'sactually launched and running and
a business model.
Hannah Clayton-Lan (33:33):
I love that.
That sounds way moreinteresting than the medieval French
that I studied in university.
That sounds great.
And so I think you'rebuilding to telling me that since
ago, your attitude changedon that, right?
Like I heard in the roundthat one of the latest Claude models
has really changed the gamein terms of the quality of the
that the LLMs are giving.
Hugh Will (33:52):
Yeah, it's changed so much and in a way that I just did
predict.
So I'm starting to think,Hannah, that coding by hand may be
a thing of the past.
We may not need to dothat anymore.
That's big.
Yeah, and certainly I'mstruggling to find cases where
like clawed code can't dothe work.
I even heard from afriend the other day, maybe this is
(34:13):
down the down the mill,but that Microsoft is now mandated
every software engineerneeds to use clawed code who works
Microsoft.
Hann (34:21):
Well, I mean, if it's that good and it enhances productivity,
that makes sense.
But then is this sort ofdarker undertone as that means you
fewer engineers, right?
Hugh W (34:30):
Yeah, well, the question is what's software engineering,
And I think I like analogies.
We use a lot of thosethroughout the show.
But what I would say isit's a little bit like power tools
the carpenter.
The average carpenter'sgot a nail gun, you know, they've
an electric saw, thesekinds of things.
They plug batteries intothings, things are just a lot faster
and a lot easier, andthey can do a more professional job.
And I would say thingslike Claude Code are a little bit
(34:51):
power tools for the carpenter.
But I'd say that softwareengineering is even broader than
It's more like buildingsoftware.
Hannah Clayton-Lan (34:56):
Yes, we did.
Yeah, we talked a littleabout this, like the architecture,
plans.
Hugh Williams (35:00):
Yeah.
So I think do you needsoftware engineers?
Absolutely, you do.
But will they be coding as much?
Probably not.
And so then I guess thequestion becomes do we need as many
software engineers giventhey spent so much time coding?
So I think I think it'syet to play out.
I think probably theanswer is that the best software
are going to be paid evenmore and uh going to be very, very
valuable.
But I think that bandthat's not so great today, I think
(35:23):
know, could be in alittle bit of trouble.
Hannah Cla (35:24):
And doesn't what you described potentially blow the line
product management andengineering even more?
Hugh Williams (35:31):
Yeah, absolutely.
And I think if I wasgiving advice to a young person
I'd say become a productmanager who understands data science
and vibe code like crazy.
Hannah Clayton-Langton (35:39):
Yeah.
Before LLMs entered thescene, there was a lot of companies
were like offshoringengineering capability to low-cost
I think like India is themain one that I know of.
I actually have a friend,actually, he's a listener who is a
engineer and he works witha team in India and he has teams in
the UK, and he says hegives directs different types of
(36:00):
the teams because theIndian engineers have like less
but they can just churnstuff out really quickly, which is
helpful.
But do you see that beingdisrupted by the LLMs?
Hugh Will (36:10):
I think massively so.
These folks who work, youknow, in these remote geographies
given really specifictasks really are mostly just coding.
Hannah Clayton-Langton (36:17):
Yeah.
Hugh Williams (36:17):
Right.
So they're not reallydoing the whole software engineering
They're just asked to fixthis piece of code, write this
whatever else it is.
And that can prettyeasily be done by tools like clawed
today.
I think that'sdramatically going to change.
Hannah Clayton-Langton (36:31):
Yeah.
Just to be clear, it's notbecause they can't do that software
engineering piece.
It's just that the way techcompanies have set up their offshore
arms is that they give veryspecific like to-do lists to people
who aren't as embedded inthe product.
And so, yeah, it soundslike that's that's the thing that
end up pretty disrupted.
Or in I mean, I guess youcould say it's enhanced, right?
You'll get a lot moreproductivity, but it just means that
(36:51):
will be fewer jobs togenerate the same amount of output.
Hugh Will (36:54):
Yeah, or there's more companies created, more happening in
and those people aredeployed into those, into those new
and new spaces.
Hannah Clay (37:03):
So we'll we'll see.
Tell you what though, youhear these stories where I think,
I think I know some people,this has actually been true.
People have more than one job.
So if you're in one ofthese sort of remote engineering
there's like a good historyof it's probably not just
but it's probably an easyone to sort of fake that they'll
more than one job andthey'll just be like pretty
and sort of be able tofudge doing two different jobs at
(37:25):
time.
And uh yeah, I guess onceyou have this enhanced productivity
expectation, it might beharder to fudge that through.
Hugh Williams (37:32):
Yeah, absolutely.
And I'm I'm sure there'salready people uh lying in hammocks
across the world withlots of agents running, doing all
coding for them andsending bills to people.
Hann (37:40):
Yeah, I mean, I think it's kind of fair enough if you can get
away with it.
Maybe that's a bit morallygray, but like at work, my line is
kind of like if I can'ttell that you've used Chat GPT or
similar, then I actuallydon't really mind.
It's just that when theoutput is it, what is it, the M-dash
that you is a real give.
Like if you can tellsomeone's use it, then they've not
well enough.
But if you can build codeand bill people for it in a way
(38:01):
you know, you're gettinggreat return on your time in the
I think that's fair enoughuntil, you know, why does it matter
if the person gets the codethat they need at a price that they
think is fair?
Hugh Williams (38:09):
Yeah, absolutely.
And I think, look, LLMsare great for two things.
You know, one is to sortof stimulate the thinking in the
place before you actuallydo the thinking and then get on with
the task.
Or, you know, if you'vedone all the thinking, you've
it through, and you justwant to, you know, correct the
write that as aparagraph, turn those bullets into a
whatever it is, then, youknow, power to you.
Hannah Clay (38:28):
Yeah, power to you.
Okay.
And I feel like, is it fairto say that the cadence of change
in Silicon Valley or bigtech in general is just getting
and quicker as we gobecause the world just moves faster
Hugh Williams (38:40):
Yeah, I think so.
I mean, I think uh, youknow, AI is super disruptive, LLMs
super disruptive.
And so there's certainlybeen unprecedented change in the
few years after what Ithink was a period of lull.
But if you look back farenough, you know, we had mobile, the
smartphone, that was apretty big revolution.
We had the personalcomputer was a pretty big
So cloud computing wasarguably as big a revolution, though
(39:01):
a little bit slower on the boil.
But certainly we'vealways had periods of revolution
by periods of stability,but we're certainly going through
of those periods ofrevolution right now.
Hannah Clayton-Lan (39:09):
Interesting.
So don't retrain as asoftware engineer just yet.
Get into data science.
Um, okay, so let's justtake a big step back out to round
this episode, which is whydoes big tech matter?
you're not interested inbeing a part of this industry, then
it just sounds like a bunchof young people making more money
than they have sense todeal with.
But um, there's some prettysignificant geopolitical impacts
(39:30):
days from big tech, right?
So the obvious one is ifTaiwan's invaded by China, that
we mentioned up front, youknow, that could cause some pretty
significant disruption totech supply chains, which will
everyone's life, by theway, guys, because everything is
by tech now.
But also just things likethe latest US administration's
to, well, basically everything.
But I've heard some sort ofEuropean tech leaders casting some
(39:54):
doubt over whether theyshould be using the big cloud
providers out in the USbecause they just don't know what's
to happen.
Hugh Wil (40:00):
A lot of conversations right now about sovereign AI
So, you know, a lot ofcountries thinking, hey, we should
our own AI capabilitythat's totally walled within our
so that we control thisinfrastructure, we control what it
do, we can make sure thatthe laws that we have are applied,
we can make sure that ourdata stays in our country.
So certainly lots ofdiscussions about that right now.
(40:22):
And I think governmentsand institutions are getting more
more uncomfortable withthe idea that their data could be
anywhere in the world andyou know, stored even in the US.
Hann (40:31):
Yeah, I think I saw France the other day were the latest
to announce like a for themwhat was a significant investment
in AI.
Although I think thecontext in which it came up, someone
saying, yeah, it sort ofpales in comparison to what these
US tech companies are making.
And I guess that's where itbecomes difficult to establish
If you're a country thatwants to play in this pond, right,
I'm sure everyone does forreasons we've mentioned, if you're
(40:54):
not well established withthe capability or the investment in
it, you're startingbasically on the back foot.
Hugh (41:01):
You need the intellectual capital, you need the the actual
to really invest, youneed to go and find people and get
to stay, have the rightregulatory environments, all these
of things, the datacenter infrastructure.
So, you know, very, veryhard to replicate something like
Coast tech, you know,anywhere in the world.
And and indeed, you know,if you're going back to the
Seven means a good way toround out the show, the market cap
(41:23):
of each one of those ismore than the GDP of, you know, most
nations on Earth.
Hannah (41:28):
Yeah, it's pretty crazy.
Like they make up 30% ofthe SP 500, which on a global scale
you know, outsized again.
And anyone that's investingin like an index fund, you know, if
you've chucking yourpension or some of your savings into
tracker or something likethat, you're going to be investing
these companies.
So you kind of can't getaway from them.
Hugh Williams (41:46):
Absolutely.
Hannah Clayton (41:47):
All right, guys.
Well, apologies if any ofyou listened to that on your morning
commute and you'rewondering if you've made the right
Hugh Williams (41:54):
But thank you for listening.
And we love, we love allof our listeners out there who
to the Tech Overflow podcast.
You can find us on socials.
We are on LinkedIn, we'reon X, we're on Instagram, we're also
on TikTok.
Hannah (42:08):
We are on TikTok, guys.
Time to get an account.
Hugh Willia (42:10):
And YouTube Shorts.
So you can find us allover the web and if you want the old
school version,techoverflowpodcast.com.
And we're of courseavailable on all of the podcast
Han (42:19):
And if you like what you've heard today, do recommend the
to your friends and family.
Give us a like, subscribe,and leave us a review.
We love to hear from you.
Hugh Williams (42:28):
Yeah, absolutely.
Go to Apple Podcasts,write a review if you really want to
thanks for the podcast.
Han (42:33):
And guys, you heard it here first.
It's time to retrain as adata scientist.
So I better go and lookthat up now.
Hugh Williams (42:37):
Yeah, awesome.
I'll help you, Hannah.
Hannah Clayton-Langton (42:39):
Awesome.
Let's go and do that.
Thanks, guys.
See you next week.
Hugh Williams (42:41):
See you soon.
Bye.
Popular Podcasts
Hey Jonas!
Hey Jonas! The official Jonas Brothers podcast. Hosted by Kevin, Joe, and Nick Jonas. It’s the Jonas Brothers you know... musicians, actors, and well, yes, brothers. Now, they’re sharing another side of themselves in the playful, intimate, and irreverent way only they can. Spend time with the Jonas Brothers here and stay a little bit longer for deep conversations like never before.
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com