Near Future Podcast

Tom and Jonny sit down with James Storer, who leads design in Monzo's AI and personalisation collective and is the biggest token burner in the whole company. They dig into the fake Monzo his team built to prototype with, why he stepped off the leadership track to get back in the tools, and what he went to Config to ask Figma.

CHAPTERS
  • 00:00 Introducing James Storer
  • 02:12 Why go back to doing the work after leading teams
  • 10:13 The founder's fight: ship it, or make it better
  • 14:47 Being a loud mouth turned out to be the skill
  • 18:35 AI lowers the floor, so who raises the ceiling
  • 25:30 Everyone an expert in everything, or still not enough hours
  • 28:01 Building hardware when you can't solder
  • 30:07 Should there be an app store with one user
  • 33:43 Is the value the software, or making it with your kid
  • 36:08 The UNO clone his six-year-old wanted to explode
  • 40:01 A 69-year-old vicar and the church website
  • 41:31 The messy middle: AI inside a regulated bank
  • 43:44 A fake Monzo with fake money
  • 49:36 Handoff should be a handshake
  • 54:16 "What will you do to make me open you every day?"
  • 60:39 Quickfire
  • 68:31 Wrap
PEOPLE
COMPANIES
TOOLS AND PRODUCTS MENTIONED
ALSO REFERENCED
NEAR FUTURE

Creators and Guests

Host
Jonny Burch
Co-founder of Near Future
Host
Tom Harman
Co-founder of Near Future

What is Near Future Podcast?

A regular podcast covering design and AI from the founders of Near Future, a boutique AI consultancy focused on teams that care about craft. We cover both what we're seeing on the ground and industry trends, ways of working and occasional guests from the design world.

Jonny: Hi everyone.

Welcome to another episode of the Near
Future podcast with Tom and Jonny.

Today we have a guest, a brilliant
guest, a guest I know very

well, a guy called James Storer,
who, uh, is currently at Monzo.

We probably a lot of the time, uh, this
episode we're gonna talk about, you're up

to at Monzo with design and AI, James bi.

I'll give you a, a quick, uh,
bio because James and I worked

together back in the day at
Deliveroo, um, sort of 20 15, 20 16.

James then went on to be head of design
at Freetrade, uh, which some people may

be using, and then went onto leadership
at Habito and then Klarna, and then

started his own, uh, design agency
Twist, and then became a co-founder of

a startup at Jitty, uh, which is, um,
a house buying and, uh, room scrolling,

I believe was one of your, one of your,
your words that you invented there.

Um, James and then has finally ended up at
Monzo where, and correct me if I'm wrong,

James, but you are the, uh, leading token
burner in the design team, uh, is in the,

in the whole organisation, though shortly,
not in the design team, um, at Monzo,

which is an absurd claim, um, for a, a,
a company that has thousands of employees

and really the engineers should be winning
that, but somehow you've snuck in and

I think we're gonna, we're gonna dig
into that, uh, during this conversation.

Uh, but great to have you here, James.

Uh, I would love to, uh, start the, the
questioning, um, the interrogation, uh,

by asking you, uh, by asking you a little
bit about, um, your day-to-day at Monzo,

like what you're up to right now, um, how
you think about your work and, um, and,

and also I suppose I'd love to, for you
to touch on your decision to go back from

leadership and like, you know, running
design teams and, and even building whole

companies back into an IC role at Monzo.

I think that's really interesting as well.

So I'll leave it.

I'll hand over to you, James.

James: Thanks Jonny.

Appreciate the intro.

Um, yeah, it is actually the, yeah, the
whole whole org that I'm lead leading

token, uh, burner, which is, I'm not
sure if that's a great, uh, title to

have, but it certainly says something
about, as you asked, like, how am I

spending my, my day to day, I guess.

Um, but yeah, I, I, I think I
am in quite a unique position.

I'm very, very fortunate position
actually because, so I lead design

in our kind of, let's say, AI forward
and personalization collective, where,

you know, you might start to see
some features coming about from in

the next, uh, next 12 months or so.

And I think what gives me a really great
remit there is because, a, we are very

much thinking, uh, more forward about
Monzo and what the product's gonna

look and feel like over, over time,
especially around personalization.

Um, you know, you and I today probably
have a very similar looking app,

and we want to make that way more
personalised for each, each individual.

Um, so I get to kind of really think
more broadly and ahead of time.

Equally, I'm not in any
particular dedicated squad.

I'm leading across the collective, so
I don't have that kind of day to day,

uh, you know, not that all designers are
like on demand to a squad, but that, you

know, you are, you are still working with
the squad to hit certain goals, certain

features that you're building towards.

I don't necessarily have that
specifically I do at the minute.

I'm covering, uh, uh, two sabbaticals
at once, which is not, not ideal, but,

uh, I hope you and Alex, if you're
listening, you're having a lovely time.

Jonny: Yeah.

James: and yeah, I, I, so I
am actually getting into the

weeds a little bit right now.

She's also very nice.

Um, but aside from that as well.

also own quite a lot of our like
dis design discipline AI goals.

So two things go hand in
hand really nicely for me.

So one is obviously how am
I applying it in my work?

How am I personally out what
we should use for what purpose?

And then to the whole team is like,
okay, what have I found useful?

How can we kind of disseminate
this through the whole design

discipline and, and beyond really?

Uh, and I'm not the only person doing
that, but it's just that obviously

I'm, I'm kind of owning a lot of those
goals, so I'm trying to figure this

out and as I said, in that kind of
great position of to really, uh, you

know, do what you say, put your money
where your mouth is, if you will, I

get to kind of do that day in, day out.

Um, so I, I would say that I spend a
lot of my time playing around in Cursor

in Claude Code, uh, whether that's
prototyping or figuring out we can use

it to, you know, build data dashboards
really easily or interrogate insights or

all of these things that I, I think, you
know, one time or another are all part

of our, our roles as designers, but when
weren't necessarily quite successful,

weren't, weren't so easy just to do.

so it's a really great place to be.

Well, what are time to be alive?

Um, and I, and I actually think
that probably answers the, the

third part of that question is
like, why, to go back into IC work.

I think I've actually, I mean, you,
you probably know this, if you looked

at my LinkedIn for, for a little while
this morning and maybe Tom's Claude

skill didn't pick, pick up on this
one, but I've kind of flip flopped

between management and IC work, uh,
like to the last 10 years or so.

uh, predominantly then, you know,
you'll know this as well as I do,

it was the only direction you had
to become a manager, a lead to be

able to, to, to sort of progress.

So I did that.

Then I did the, the kind of classic I,
I'll I'll manage some people and then, you

know, you realise that I'm, I'm actually,
uh, the best way of putting this?

I, I'm a great leader.

I think I'm a good lead by
doing, by showing how to, to,

to sort of lead from the front.

I'm not sure I'm the best manager.

I've, uh, I've, I've tried to a few of
my, I mean, I've got a few reports now.

I'll, I'll let them, uh, sort of
hang, hang me out to dry here.

But I, I, I think I, I described
myself to, to one of them

as a bit of a chaos agent.

So I think, I think it's not always
the best thing, but I, um, I really

love helping people progress, but I
really love doing, doing the do as well.

And what's, what's worked
really well for me there.

And as I've gone through these cycles
of leading a team and then being

like, oh, I just wanna get involved.

You sort of then have to go
back into an IC specific role.

I think what's happened really nice there
over the last 10 years is that there's

been these more hybrid roles come about.

You have kind of, you know, senior leads
and all this kind of stuff now, um, and

more hybrid, uh, sort of player coaches.

So, so that's worked really well for me.

And you do kind of gain a lot
of, uh, you know, a experience,

b insight, understanding of
how to play the strategy game.

You get all of that and when you can
package that up with a, you know, a 20

year career of, of, of being a designer,
you can do some great design work as well.

So I, I think that for me has been,
uh, the, the kind of pinnacle and

be, but going from like a founder
position that that is, you know, people

might see that slightly differently.

I'm, you know, for anyone that's done
it, John, you'll know this for, for

sure, you are, you are in the tools like
24 7 because there isn't anyone else.

There is no team.

You are it, you are the first
and last line of defence.

So for three years, you know, I was the
only designer that's got some benefits.

You get very, very into the tools.

Uh, but the, the, but the side part
that made me want to go join in a

more established team was peers,
having people to learn from, to

continue to develop myself as well.

And equally just, you know, having some
really, really great people to work with.

Uh, again, I think Tom, you'll have
worked with some of, uh, the, the

great folks over at Monzo that, you
know, what a fantastic team it is.

Tom: Is a fantastic team and you're,
you're lucky to work with them.

Um, love that story.

The one thing about the career back and
forth that I was particularly curious

to ask you around, and I'm curious with
this AI shaped role that you're doing

at the moment, do you see that as, it
seems like this sort of sweet spot of,

of bringing together your IC focus and
there's some people leadership and there's

some direction setting across the company.

I'm curious if you see that as like
a, what the time horizon is for that?

Is that like a one year role that then you
switch back into IC or people management?

Or do you think the future of this craft
is this hybrid together of, of, uh,

disciplines through AI enabling that?

James: A really great question.

Uh, ho I don't know for sure.

I'm not sure any of us can know where
the end state is right now, but I,

I, I, I don't want to, uh, to lay too
heavily one way or the other, but I,

I think that there's definitely, uh,
always gonna be a need for people that

are very strategically minded, great
leaders, great managers, like people will

always need that to be able to develop.

And, but I do think like, just like ICs
are having to expand their skillset,

their breadth, I think that's kind
of the same in terms of leadership.

Like they can't just turn a blind eye
and not know what's happening in AI,

the tools that we're using, you know,
but I, I, I spent enough time like

out of Figma that going back into it
was hard at the time and they didn't

even move that quickly at that point.

But I think that now you can't
really take your foot off, so

your eye off the ball really.

So I think it's, I think it's,
they're always gonna exist.

I'm not sure whether the, the, the hybrid
role will be there forever because I

am, I think for me personally, as much
joy as I take from helping people,

managing people, I'm probably better
at, at actually just doing the work.

Um, felt like, uh, the current
collective that I'm in needed somebody

closer to the actual problems.

The work that was happening
at the time when we, we sort

of started the collective.

I think that made a lot of sense.

I dunno, going forwards, once we actually
establish ourselves, start to ship some of

this, this work, um, I'm not sure whether
that might then need somebody to be a

little bit more sort of just day in, day
out in the actual, in those leadership

meetings in the in Australia sessions.

I, I try as much as humanly
possible to stay outta those.

Um, but you, you can't, you can't, you
can't sort of do that all the time.

So.

Tom: Daily.

interesting.

Jonny: I, I suppose I've got another
follow up question, and this is

from personal experience as well.

As a founder, as you mentioned,
James, when you're a founder, you

don't have the designer title,
but if you're the design founder,

then you are the designer as well.

So you,

James: Yeah.

Jonny: I think I found myself in a
situation where I was doing a lot of

design work, but always struggling
to justify spending longer than I

needed to on it because I was, as
I was CEO and, and also doing like

fundraising at various points and I,
and product management for the most

part, for, uh, for long periods of time.

And so the PM bit of me would be like,
what is the smallest thing we can get out?

And then the designer would be
like, but I wanna spend a bit longer

to, to, to make it a bit better.

And then the, the PM help would come back
on and be like, no, ship it, move on.

Um, so I'd end up like, looking back, I
think the ways that I sort of tried to

combat it was, you know, promising myself
that I'd go back and improve it over time.

And I ended up designing way more in code.

'cause it was just faster to get to
something that was high fidelity anyway.

But I'm just interested, like, as you
left being you one of three founders,

but like, you know, various things
on your plate and lots of competing

priorities as a, as a founder of a
early stage startup, then going into a

big team, you mentioned being out the
tools for a while is hard to catch up.

Was that a moment where you had to
sort of recalibrate how long you

could spend on craft or how, how you
would sort of reprioritize where,

where to where to put your time?

James: Yeah, I mean, I think like one, one
of the, one of the driving factors for me

to, to sort of leave founder life was that
many hats and at points it's enjoyable,

it's great, the energy's there, it's
fun, but it can also be an energy sort

of snap it can take figure out of you.

And so I think for me it was more
around like, yeah, being able to have

that extra time to spend thinking
about something you want to, um, I

just needed that mental energy to
have that, that ability to do that.

And do me wrong.

I think AI has been phenomenal for
being that thought partner, being

able to like accelerate parts of
your work, like you said there, you

know, have resonated a lot with me.

I was predominantly like building
features by the end of Jitty.

I, I wasn't really designing everything
first because one, we built a really

great design system, so you had all
the Legos available and you know, you

can't always be too precious in a,
in a startup where, you know, days

and hours are, are, are your enemy,
not the, the weeks or months it might

take to ship features in bigger orgs.

So you sometimes you just
have to have to deliver.

Um, and so that, that was definitely
thing I really wanted to gain from sort

of like moving on from founder life was,
yeah, just having the headspace to go,

my priority today or this week is to
literally think about this problem or go

and build a, a prototype in this space.

And,

Jonny: Right.

James: you know, it's happening right now
at the minute we we're doing a lot of, um,

our cycle planning for the next half and
I get to go and kind of prototype a lot

of those bets because that's become part
of our process now, which is fantastic.

But it's, I mean, so for me it's, I
I sound giddy saying it like, it's

like it's both great to get the time
and head space to go and think about

these like, future problems we wanna
solve, but equally getting to, to play

around is the better way of putting it.

It's to play around building these things,
uh, not, and you know, the, this is kind

of what we touched on last time, Tom, is
that not to the level of sort of fidelity

or quality execution where I would be
shipping them, but to the point where we

can share them round as artefacts, share
them round as like, uh, you know, sort

of direction of travel, and prototype
them in enough to test with users.

That's what's really, really created
some fantastic leverage for us, uh, at

Monzo so far and, and given me this,
the giddy awkward smile that you see.

Jonny: I know you James, and I know
how you work, or at least how you

worked 10 years ago and uh, I can
only imagine how exciting it, you, you

were in the growth team at Deliveroo

James: Yes.

Jonny: in the day, which is was
the team that moved the fastest

and had the ideas and then just
like would throw them out there.

I think you and I are quite
similar in that sense.

Uh, just like to sort of build something
and just see, so to be able to have,

uh, an organisation that operates where
as part of planning, not only do you

get like maybe an engineering spike
into some sort of technical question,

but you actually get to build like a
sort of user experience spike as well.

Man, we would've loved that 10
years ago, I think to, to have

had that opportunity to do that.

James: yeah, I mean, I, I, I say it
quite often at the minute, but I,

I, I feel like, uh, you know, I've,
I've been a, an average designer for

the most of my career, and a lot of
the, what, the many things that I've

done, I, I, some people don't know.

I, I started as a, as a web
designer, as a web engineer back

in the early two thousands and
sort of moved into this direction.

But it's a lot of the skills that
I have acquired over this time, I

feel like they've come to a head
where it's like, I do understand, uh,

uh, a lot of the coding decisions.

I mean, not to the amount of
coding languages that I now,

you know, orchestrate, but I
understand, uh, some of it.

And I think that's proven
really, really powerful.

This, that tech, that technical
connection, the, the problem-solving

perspective and the actual
like visual craft element.

And then equally, I think you, you
once said this to me, Jonny, it stuck

with me for a long time, is that, you
know, I, I, I used to call myself loud

mouth and I thought it was a negative.

And you said it's actually a, it is a
superpower, it's a communication skill.

thanks for that.

And, uh, and I think that that coupled
with these other things has become

this way of like being able to be on
the forefront, forefront of, of AI

and design and understanding how we
use it and helping others to kind of

come along on that journey as well.

So it's a, it's a, it's a weird collection
of skills, but seem, it seems to be

working pretty well for me at the minute.

Jonny: Yeah.

And you, and I love that you've
managed to just sort of define

yourself between the cracks of like
the, the traditional archetypes.

Uh, you, I think you are the best example
so far that we've had on this podcast

of someone who is doing exactly that.

They're sort of dancing between these
definitions and actually carving something

out for yourself that is, um, the best
reflection of what you want to do.

So therefore, probably the best value you
could, you can offer to an organisation,

James: Yeah.

Jonny: which is great and probably not
possible five, 10 years ago for various

other reasons, like, you know, coding
models didn't exist and, and whatnot.

But, uh, but it's, it feels like it's
breaking down those silos and I spent

years thinking about career paths.

I'm so glad that, um, we don't have to
always talk about the management track in

the, the IC track, in, in that same way.

James: You should, you should go again.

But with all these little,
uh, micro, uh, uh, tracks,

Jonny: Thank you, James, but no way.

Absolutely no way.

James: the, the, the interesting
you mentioned archetypes there.

We, we, we had a, um, we had like a,
a senior IC, uh, session with, with

some designers a couple of weeks ago,
and we started talking about different

archetypes of our most senior ICs
and like what makes us all unique?

What are our superpowers?

How can we bring them to the groups,
uh, to the, to the wider group?

And, and, you know, everyone does
have a different path and it's now

starting to sort of like, where,
where are your sort of like skills,

your strengths, and how can you ex
sort of amplify them with AI, um, and

with your peers around you as well.

my, the, the one that kind of came out
for me was, uh, was the trellis, which

I'm not, I'm not sure I, uh, appreciated
so much, but it was essentially, uh, you

know, we had some folks that were more
around like, uh, you know, being like a

sturdy founding, uh, foundation or like a,
a holding wall and they might be like more

systems focused and mine was more around
like, I enable others to kind of climb

because I, I'm a wooden fence apparently.

But, but I think I took it as more like
as, as a, as a compliment, it's like more

of an enablement, uh, skill set than,
than me just being a, a, a wooden trellis.

Jonny: An uncanny resemblance.

James: Thank you.

I appreciate that.

Tom: Um, and for anyone who's just
tuning in, we were, uh, talking about

walls before we started recording,
and James', uh, taking, taking a long

time to knock down various walls.

And I feel like that's come full
circle now with the, with the

James: Yeah.

Tom: the wooden fence.

Um, James, I had a question for you.

Uh, I think it was a, a previous
podcast guest, Avi, at Deel, um, he,

um, posted this morning or over the
weekend on LinkedIn about the idea of,

um, raising the floor and raising the
ceiling happening in different ways.

And the, perhaps the more traditional
model, and this is my interpretation

of what he was writing, but the more
traditional model of outlining what

skills are required for a team and
the sort of core competencies of a

progression framework is that floor.

But the ceiling is kind of this
more exciting, uncharted territory.

Um, and your comments there about,
um, archetypes and specialisms, uh,

sort of start to get at that a bit.

And I remember from my time at Monzo as
well, there was this sense of once you

reach a certain level, say staff designer,
everything above that level, if you're

looking for a promotion or looking to
get more senior in your, your role really

needs to come from you and really be
driven and defined by who you are and

what you are, what, what value you're
bringing to the, the organisation that

is uniquely in terms of how you operate.

I'm kind of curious with your sort of
theme here, how much of it is about

AI at all versus just individual
growth and sort of skill improvement?

Like, is there a specific AI lens to it or
it's just like, AI exists now, so why not?

Like maybe we should talk about
this and, um, and there's probably

AI ways AI can help with all of it.

Or is there a more structured way that
AI is sort of integrated into this?

Because I'm curious how, how that does
raise the ceiling of the team, um, and

with an organisation of a hundred or so
designers that has the power to be very

impactful at setting or raising that
bar for, for what teams are capable of.

James: Yeah, it's, it's a good question.

It ca it came up, I, I went to Config
this year and it, there was a similar

quote about sort of AI lowering the
floor, but it's, you know, it's up to

us as designers as, as, as technical
and non-technical people to, to raise

what the ceiling raise, what's possible.

And I think this, this comes across
for me in terms of what I look for

when I'm thinking about kind of craft
now, when I think about design craft.

and just to be clear as well, we,
we haven't like solidified anything

in our progression framework yet.

I think it's a little too early.

The, we are gonna make some changes
around just, you know, staying,

uh, curious, continually learning.

Uh, and I think AI fits within that band.

again, it's like, you know,
you don't wanna brush AI on

everything just for the sake of it.

I think it's trying to understand that
actually what are we looking for and,

and how is that gonna sort of help
both individuals and the wider team at

Monzo, uh, sort of elevate and that, and
I think that's, that's the key for me

is that, you know, we've always looked
for, in product designs, you always

look for strong breadth of skills, uh,
like craft skills, um, amongst others.

And then you look for those
like individual depths, the,

the, the, the T shape, right?

So I'm talking to two, two very senior
designers, apologies, and one that built

the book on progression frameworks.

So you see what t is,
Jonny, you, you take this.

But, so the, the, the point I'm making is
that I think there's, there's this a lot

of information at the minute that makes
it sound like just expected to expand

and expand and expand our role, and AI's
gonna let us do all of these things,

and all of a sudden we're gonna, we're
gonna be product managers and engineers,

and we're all gonna just keep going.

I'm not sure I view it quite as the same,
but I think it, it widens what we are able

to do and what we're able to do better.

Like, there's been a bunch of times
in my career where, you know, uh,

data has not been a strength of mine.

Or like, motion, for instance, skills
that you, you know, are definitely

on that craft, uh, sort of framework.

But you, you don't always expect,
you're not expect to have them

or, or even have a depth in them.

I think that the, that's the floor
sort of lowering to me is that I now

can go onto Figma and I can say, can
you make these shapes move please?

And it, you know, it might, it might
do it maybe, but, but the po the real

point I'm making is, is that it allows
me to have this base entry at the very

least, to where we have incredible
motion designs, by the way, at Monzo.

And I wouldn't belittle their skills,
but that's their depth, right?

And so where I look at it is,
is that it allows me to do that.

It allows me to interrogate
BigQuery or look at, um, NotebookLM

Notion and look at our latest, uh,
research studies and really try

to pull some snippets of insight.

It shouldn't replace us going and
doing research in specific areas,

but it's great to be able to
quickly pull together something.

So that's the breadth for me where
I think that AI really is gonna

accelerate people is the superpowers.

It's like, how do you take a
strength that somebody already

has and it into a superpower?

It's like, you know, don't, can't
think of an example particularly, but

I was gonna go for some kind of like
Thor example, but he's got a hammer.

So, but he, he, but I think, I think
the point is, is that if you're good at

something, you can be even better at it.

You can accelerate that.

And I think that's, like, for me it's
the, it's the, it's the technical side.

It's the being able to build
stuff, it's the understanding.

And I've always done a good job at being
able to connect those worlds and really

being pragmatic about what was possible.

And I think for a lot of people
now, they can, like, even at the,

at the base level of that, they
can understand that a bit more.

Because if they try and mess around in
code for a few hours and you can't, if

you can't make it do the thing you want it
to do, it's probably not that easy to do.

So, whereas if you have an interaction
in mind or you've seen it and you

want to sort of play around with it,
you could drop a video of it in, we

do this a lot now, you sort of like
reverse engineer interaction patterns.

think.

Like those things for me are
like, they are really being able

to accelerate those superpowers.

And so that was a very long answer
to your question, Tom, but I think

that's how I see the difference
between sort of like the lowering the

floor and the raising the ceiling.

Tom: Yeah, I really

James: I'm not sure that's the answer
to your question, but I, it's an answer.

Tom: It's, it's a good,
it's a good answer.

Yeah.

I think it makes a lot of sense.

And then I guess as you get more
senior, that expectation of how you

raise that ceiling is on you to know
yourself and what, what that shape

looks like, and maybe the role of
managers to help you back what that is.

Um, and I guess use AI as well to
even help reveal that to yourself,

not just to get things done.

James: Yeah, I, I mean, honestly, I
I, I used AI a lot last year when I

was thinking about what's the type
of role I want post founder and,

you know, you just talk to it and
say, these are the things I enjoyed.

These are the things I didn't enjoy.

Here's where I think my
strengths are, blah, blah.

And it can just be a really great
sparring partner in that respect.

And, uh, it's either, it's either talk
to ChatGPT or talk to my wife and I,

I, I get pretty much the, uh, opposite
I get sycophant and, no, just do it.

I'll let you decide which is which.

Tom: Amazing.

Um, James, I'm gonna switch
gears a little bit unless you

have any more follow ups, Jonny.

Jonny: No, I, I mean, it has made
me think quite a lot actually

about this sort of choice.

Is there this choice that we start
to have now where we can go and

specialise in some new direction
because we've got extra tools.

Like I could go and become a
design engineer even though

I've never been an engineer.

Maybe to maybe a little bit.

But I think what we've always found,
and we've talked about it on this

podcast before, is in any given
domain, everyone outside that domain

is like, oh, AI can do that now.

So, you know, design is cooked
or et cetera, et cetera, all

of this ridiculous stuff.

Anyone who's inside that domain and
knows what good looks like, knows

that that not to be true, and there
still needs to be true expertise.

And actually the what it means to be an
expert changes and will, will rise as, um,

the technology enables more of the, the
easy stuff or the rote work to be done.

And so, uh, it would be easy to like
tell a story of, well, everyone,

every designer can now do everything
because there's AI to do it.

But in fact, the, the standard and the
expectation is shifting such that, um, you

know, I, it, it's, it's still a function
of how much time someone has in the day.

Like, I can't be an excellent PM and
an excellent engineer and an excellent

designer because there just isn't
enough time and I don't have enough

time to get good at all of those things.

So I'll end up still being mediocre
at all of them, which is why it

would've been five years ago as well.

Um, it's just interesting to think about,
I think like this illusion of all powerful

designers, when I suspect that will not
be the end result, we will still need

to have someone who's exceptional at
emotion if we really care about that.

There'll be a set of companies
where they're like, yeah, it's

fine, you know, it's good enough.

Um, or they don't wanna spend the money
on the, the specialisms, but yeah.

James: I, I, I agree.

I, I, I don't think, the way, any
of the, the sort of AI assisted

coding or products that I do, I think
any of that makes me an engineer.

And similarly, I, you know, you said like,
what, what other things did it apply to?

Like, I've been messing around
with hardware, I've never touched

hardware in, in, in my life.

And now I'm like, oh, am I a,
am I a physical engineer now?

You know, cool.

Uh, I'm, I'm even looking at, I was
looking at soldering irons the other day.

I don't even how to use it.

I haven't used it since I was like 15.

very

Jonny: But now you work in
robotics, James, obviously.

James: a

Jonny: Yeah.

James: engineer.

Jonny: Yeah.

James: I'm, I'm, I'm, I've got promoted.

But I, I think, I think that,
hon, honestly, you are right.

None of these things
makes you those things.

It doesn't make me a great engineer
or a great product manager, et cetera.

And likewise, it doesn't make an engineer
a great designer if they go and prompt

Stitch to, to throw a few visuals out.

It's, it's still the core skills that we
have make us great at those specialisms.

But I, I love the mentality
of I can go and try things.

I can do, do X, Y, and Z.

Um, and I would love
to, to do more of that.

Like gen like the hardware thing came
about from, you know, I've, I've,

I've conquered all these mountains,
like what's, what's, what's next?

Um, it was actually 'cause I wanted
to build something for my kids.

But I, I think, I think the, the great
thing is, is just that, that feeling of

possibility of like, that you can have a
go at anything where previously, you know,

the first hurdle that you hit and you're
like, nah, I'll go back to drawing boxes.

Tom: Yeah.

What, what have you built recently?

What?

What are some examples of that hardware?

James: Yeah, so we, we actually had a
guy, um, Rob, uh, who had been playing

around with these little M5Stack sticks.

Uh, and he'd managed to get his kids
like balances, uh, showing on them

because obviously like without a
phone, uh, on kids accounts at the

minute they can't see their balance.

And my, my two have, my youngest
just turned six, uh, this year,

so they both now got accounts.

So I, I built that.

Um, then that kind of like goes,
oh, I can build on hardware now.

So you, you sort of, you branch out
a little and, uh, I built, uh, an

app on our tv, our Samsung tv, uh,
for like a, to show, sort of show our

calendars to show things like, uh, we
haven't got like, to-do lists a lot.

A lot of it's through like Google,
uh, you know, you'll have like

Google apps that do a lot of this.

So if you kind of just build in,
build into an ecosystem, you can

get quite a bit quite far from it.

So just showing like our our to-do
lists, like where we're gonna be and

when, like, um, and then we, I did put
the kids balances on there as well,

just so they could see it on the screen.

And it is just something like that
that's really fun and enjoyable.

And even if my wife

Jonny: I assume you didn't put
your balance on there, James.

James: no, I mean my, I, I should
just write note on the screen

and it's, uh, it'll be done.

Jonny: Hard coded.

James: Uh, yeah, e even if my wife refuses
to open it, I'm still proud that I did it.

But I, I think those things, it's like,
it is, it's like a, a bit of a gateway.

It kind of like pulls
you into the next thing.

So now, um, I dunno whether you're
familiar with the, the, the sort of, um,

the LightUp keyboards that you can get.

Uh, uh, it used to be called Lumi,
but it's, it's called ROLI now.

So they, they're, they're
unfortunately not doing very well.

I hope that they pull out of it, but
what they did was open source, all

of the, the code so that if it does
go down, then people can still kind

of use the, the hardware devices
that they've, that they've bought.

And so I was like, oh, okay.

So now I'm trying to build a, an
app to work with a Lumi device.

So it's, I don't know, I, I think these
things, like you say, have a way of, of

making you feel that things are possible.

And I genuinely think if I, if I
was gonna be around with anything,

this would be my state of mind, not
what can I build to make some money.

And I think that's where people perhaps
go a little bit wrong is that you,

you see it on, on Reddit and other
channels, it's like, you know, my app

that's doing a hundred thousand pounds
MRR and that, you're like, okay, sure.

And, and so tho those for me is like, it's
like cha chasing the dream, whereas I'm

happier that I'm just building something
to solve my own personal problem.

uh, we have had like a bit of
internal debate around that.

It's like whether or not that will
be more of a thing going forwards.

Like if you might see more personal apps,
like would you, could you even imagine

a personal app store on your phone?

Like, these are the kind of
things that I would love to see.

Um, it's just more like niche products
than what we thought, what we've

had to date, which are kinda like
bigger products that do a lot more.

Jonny: Yeah.

James: I, I, I'd be really interested
in following that thread and seeing,

seeing like really where that comes out.

Because I don't know whether it's the
thing I want or whether it's just the

thing I want right now because it's fun.

Um, you know, if it becomes a
point of like, people stop building

these bigger apps because people
are building niche ones and I

Jonny: Hmm.

James: a app to do a certain thing, I've
gotta go and build it to just to get it.

I think that would be a bit of a blocker.

But yeah.

Jonny: Yeah.

With there, there's been a, probably about
a year ago there was this, uh, iPhone app

released Wabi, WABI, which promised to be
this sort of, uh, personal app builder.

I'm not sure how well it's
doing 'cause I haven't heard

that much about it post-launch.

But there is this bet, I suppose, that
everyone will be a builder at some point.

Everyone will build their own tools.

I suspect you and us three on
this call are outliers in, in

the sort of broader ecosystem.

Well beyond design, you
know, normal, normal.

If I imagine my mother-in-law doing
something, I, I, I can't imagine, you

know, she's gonna be digging around
trying to build something anytime soon.

But if the technology gets to a place
where it just feels like an app you're

using and you're just asking a question
and then it like, makes something and, and

it says, do you wanna do that every day?

And you say yes and then suddenly
you've built one without knowing maybe

it can sort of get to that place.

I do like what you're saying
about this sort of, um,

personal like hobby
cottage, app building type.

We do stuff for fun as a family when we
might have played a board game instead.

We'll like build a thing.

Tom, I know you are doing a lot
of this kind of stuff as well.

Um, and where whether there will
be anything in between where more

people become entrepreneurs to try
and make money, or it might just

be like hobbyists and like, you
know, people takes over Warhammer

Tom: Yeah,

Jonny: that's like where.

Tom: yeah, I've, I've
been thinking about this.

Um, we are recording this on a Monday,
on Friday, we recorded the last episode

of just Jonny and I, we, and I mentioned
in there about building a, a, uh, to

play chess game for my 5-year-old son.

Um, and it was really an ex just
a chance to explore Three.js

and what it can do with
creating a 3D world.

And we had this underwater world with fish
and things playing different chess pieces.

And, um, anyway, that was
two days ago, three days ago.

And on the weekend, on Saturday,
I was like, oh, I, I, is a, looks,

this is cool as a proof of concept
experiment, but I don't think my

son's actually gonna wanna play this.

So I had cla Claude turn it back into a
2D game, make it more of a pixel art style

with like these sort of rainbow colours
and like, want some motion animation, make

these have characters, give them eyes,
make them move around, do fun things.

Um, and then we played together on
Saturday for about an hour, um, during our

youngest times, youngest one's nap time.

Anyway, the whole experience felt like,
left me with a question of is this,

is this actually is the value here
the personal software that I created?

Or is it the, the process of collaborating
with my 5-year-old to make something

that feels more fun and engaging?

And it doesn't actually
matter if that's any good.

Um, it's the process of creating
it together that this now enables.

Um, and it's sort of hard to imagine
there being a big enough value difference

between a dedicated piece of software like
Wabi versus just using Claude Code or some

other, other sort of fairly mainstream
LLM coding agent to be able to build that.

yeah, I think the, the value was, this
experience at least, was less about

the a kid's chess game and more about
collaborating to create something that

was managed to weave in the things I
was interested in helping my son learn

and creating it into a format that was
fun and exciting for him to play with.

So turning this back around to, to
you James, and the, the whole point

of this actual interview, um, yeah.

Some of those examples you gave there with
the, with the TV and hardware and trying

bank balances for kids, I'm, I'm curious
about your perspective on involving,

involving your kids in these experiments
or your wife or other sort of people in

shaping them, or if it still feels like
it's a solo, like you are the author of

this product and it's just like the, has
a very specific niche target, but it's

less about the collaboration to create it.

James: Yeah, I, I think so.

So sometimes it definitely
is the, the latter.

It's like, I will have an idea and
I just want to pursue it, but I,

yeah, I definitely used it to, to
sort of play around with my kids.

My, my wife is usually more, she have
an idea and say, I want you to build it.

And I keep being like, I, you
know, I, I would like that it's

something that we do together, which
I think we, we will eventually.

But yeah, certainly my, my,
my, my son especially has

shown a great interest in it.

I think he kind of pokes his head in
here sometimes and sees me doing stuff

and he is curious, what are you doing?

And we, similarly, Tom, we, we built a
Nuno game, um, which for copyright reasons

I'm gonna call Duo, um, or dos do surely.

Uh, and so yeah, we, um, he
basically wanted to play, uh, more

and, and it sounds really bad.

I didn't build it to not play
with my son, just to be clear.

He went to play on his own,
but also with other people.

And you, got the card
sets and he loves them.

Uh, and he's got like a few different
versions, but he basically wanted to

build it so that he could play with
his like nanny and grampy or with

his aunties and stuff like that.

Uh, and also he wanted to make
his own versions, his own rules.

He was like, I want it to explode.

I'm like, what's,
what's, what's exploding?

And so, so we built, we built the
core game and then built like a, an

invite, uh, system so he can, he can
put their number in, it'll text, text

them a link and they can play together.

built that and he absolutely loved it.

So then we started building more
ru rules, different game types,

and he has got, yeah, he's got a,
an extreme version, which does,

uh, basically cards can explode.

So, but this, I think it's, you know, I,
I love chasing down those cur curiosities

and I think that is really great.

And I, I would love, I mean, I, I've
not got there yet, but I would love for

him to be able to do that on his own.

I think maybe at all, like,
uh, you know, that Lovable or

something like that might, might be
beneficial for him in that respect.

Um, not to sort of pencil any particular
one out, but I can't see him like firing

up Claude Code anytime soon, for instance.

But I, I, I do, I would love
for him to be able to do that.

And so when I was, when we were
building it, I was trying to get him

to do the talking, so like, speak to
it, tell it what you want it to do.

Trying to understand that this is
like prompt engineering of some kind.

it is, you know, it is
like building those skills.

I think.

I mean, I mean like, you know, for,
for what it's worth as much as what

we've learned in our X amount of
years, uh, gents, I, he, he's like

a, he knows how to use his iPad more
than I ever have known how to use it,

um, to the point where he actually
mentioned lock me out of it, which I

couldn't actually figure out how he did.

they're, and they're gonna be AI
native, these kids, so they're,

they are, they're gonna fly.

I think kind of enabling those
curiosities now is great.

And I also sort of think to sort of
bring it back to the broader design

team and how we work in companies
is, it's the same thing really.

I mean, I don't, I don't sit with
many of the, uh, staff, you know,

and teach them how to build, do know.

But I do think it's like, it's stoking
that curiosity and uh, and honestly

like, you know, it, it's just having
that kind of growth mindset, that

learning mindset that's, that's required.

Um, and I don't think that means
that everybody needs to be a builder.

I think it's the same point as I
just made about me and my wife, like

I'm the builder, but she can have
ideas and wanna know how to do it.

That that curiosity at work can sometimes
be, are a bunch of people that are

enabling, laying those foundations and
creating the right environment so that

anyone can, can pursue those curiosities.

So then when they are there and they
want to do it, you need it to be simple.

You need it to be easy to access.

And so they can just go and do it.

And we, we've had some real success
with that and I, I, it's been really

nice to see people that would've, you
know, six months ago say, I'm, I'm

terrified of AI, I dunno what I'm doing.

I dunno what's going on.

To then be like building like
full flows with, with APIs, with

data call up and, and doing it
like pretty much on their own.

It's really nice to see
that progression happen.

Jonny: Yeah, I, um, I was with my
parents this weekend and I told

my dad, who's a vicar, uh, about
Lovable, and I said, you know, you

could build the new church website.

And he said, oh, the
church website's awful.

And we have a web guy,
he's a young guy, about 50.

And I was like, of that, you know, the,
the, the average age of the sort of church

team is such that the young guys are 50.

But, uh, he's like,

James: niche

Jonny: how, how long would
it take for me to build this?

And I was like, I mean, the church
website must gonna be half an hour,

probably, something like that.

You know, just point at the
old one to get the brand and

then, you know, uh, off you go.

Uh, so yet to yet I wanna watch him do it
because, you know, if my, uh, 69-year-old

father can, can navigate Lovable and build
something good, then, you know, that's a,

there's something very interesting there.

Anyway.

Right.

Let's move on because I really want
to hit, I wasn't at mostly working.

James, you, you came and spoke at
Mostly working, which is, uh, Tom's,

uh, meetup that he runs for design
and AI in London, which is excellent.

Tom: With

Jonny: came and spoke at it with
Simon from Morse, sorry, yes.

With Simon from Morse.

Um, James spoke at it.

I couldn't make that one.

I was gutted to miss it.

I would love to just hear, uh, and
I think everyone that wasn't at that

event, we would love to hear what you
spoke about, the specific project that

you were working on at Monzo that you
demoed, uh, and, um, you know, like how

transformative that's been, because I
think every team could learn from this.

So I'll, I'll hand over to you.

James: Yeah.

Yeah.

Awesome.

Thanks.

Um, yeah, so I think it was actually
similar to what I was just saying

is that it was very much how we're
trying to enable designers to

use AI, you know, inside of real,
a real team, real organisation.

Um, and the core argument that I
was making was kind of the biggest

value that we can have as designers
and, and other non-technical staff

at the minute isn't suddenly that we
ship production code and we're like

committing to the actual code base.

In some, in some companies, that
is gonna be the biggest leverage.

And so I, I think my point was not around.

Don't do that.

It was more that you've gotta
find the right leverage for your

organisation and for us it's very
much like if we can reduce friction

to the tools, we can improve the
quality, uh, and speed to learning.

And, and basically I think I called
the talk, uh, the messy middle.

And yeah, it was really like trying to
define that gap between, uh, as I've

mentioned at the start of the, the
podcast was between like AI, you know,

the public narrative of AI is replacing
everyone and anyone can ship immediately

and make a hundred thousand MRR.

Um, but versus like the actual reality
of using AI tooling inside mature

organisation, but inside a regulated
organisation, you know, I know there are

other banks, fintechs that don't even have
access to ums to, to this kind of tooling.

So it's, you know, it is
great that we have got that.

I think for us, like the, work
has really been like a mix of,

of having this like prototyping
infrastructure that I've helped build.

Um, but also like safe
access to the tools.

Like we have contracts in place with all
of these LLMs so that we're not using, you

know, they can't use it for training data.

We're not putting PII data in there.

Um, but equally like you need
infrastructure to help you settle this up.

So you have to have like security
and platform supports, uh, to

give you sort of that capability.

Um, so it is a, it is a team effort.

I'll be, honestly, everyone's sort of
bringing their bit, but we, we, like I

said, the core from my perspective was
if we can improve the kind of quality

of our prototyping, our, our learning,
we can get these like higher quality

insights just to basically, I don't
think it's even efficiency at this point.

We, we were talking about it
more as if you look at like an

efficiency by value graph, very
much at the top end of the value.

I'm not sure we're like
that much more efficient.

You probably just spend more time
like, you know, stressing the details.

This isn't a bad thing, but I think
the higher value is that we are

able to just be more sure about
the direction we're going in.

We can get this like much,
much higher quality signal.

I mean, like you consider, even a year
ago we were probably still stressing

about how do we put inputs in prototypes.

It is, it's really, it's really
mental that we've gone from having

been able to solve that for 10 years
all of a sudden, hey, we can do that.

And also here's this entire
database backend that we can

have that mi that mimics a bank.

so, so that, that's, that's just to
bring it back to the point that's,

that's actually what we've built
is this kind of clone of the Monzo

product and, and, uh, uh, an ecosystem.

So we have, um, a design prototype,
uh, built and React Native.

So it feels, it feels like the
Monzo app, but it's using all,

uh, sort of like fake data.

We can create accounts, connected
accounts, we can simulate transactions.

And what it's allowed us to do is to
be able to build much faster into,

you know, more murky directions.

You can prototype much more quickly.

You have an idea.

And rather than jump into Figma sometimes
it's much quicker, especially where a

lot of the things that we're thinking
about at the minute are AI enabled.

It's quite hard to do
that on a static canvas.

You use using a, a prompting
agent as your canvas.

You say, this is what I want it to
do, here's how I want it to work.

You might want to add some colour on that.

This is what it, I, I've sort of
like mocked it up to look like, but

we built it all on our rails of our
component system so it knows of our DS.

It's like, it, it's, you can
get much, much better quality.

I think I called it, um, previously
where you might have used

Figma Make, Lovable or similar.

You kind of got the, uh, the Temu Monzo,
which is, you know, it it, it just comes

out splashed with a bit of Hot Coral.

It looks awful.

It doesn't look anything like the product.

And I think we, we actually,
honestly, you know, I, I'm, I'm

still, I'm still believing using
the right hammer for the right nail.

And I'm not saying that the React
Native app is their only way to, to

test anything or the inner way to learn,
ultimately as well, you know, it's, it's

difficult to go out to users as a bank,
show them this hacked together product

and be like, can you, uh, can you, you
know, what would you do with your money?

Or could you tell us how, where
you would put your account details?

And you know, they're gonna
go, is this, is this real?

Is this, is this actually the case?

And so I, I think you, you know,
you have to think about it from

a security perspective as well.

Um, but yeah, what we've built, I'm,
I'm, I'm really pleased with and

yeah, it really is enabling us to,
to sort of move, move faster in,

in lots of different directions.

And, and like I said earlier, it's great
when you see people who previously were

a little bit more AI shy or a little
like unconfident using those tools up

with this whole prototype of like their,
the thing they've been working on.

And then testing it with users as well.

We had a sprint, uh, two weeks ago
on a, on a, on a product I can't talk

about, but they, they literally, like
start of the week, we just, were were,

we were jamming on ideas, we were
thinking about the specific use cases

and then they turned it into prototypes
and tested it by the end of the week

using this React Native prototype.

It was fantastic to see.

So it's, yeah, I, I, I really love
that, that level of, of quality and

uh, and craft that we can get out
of that in, in such a short space

of time is, is just phenomenal.

Jonny: I do remember the, uh, we
used to do design sprints together

in, out of the Sprint book.

And just think the Sprint book now
is probably, you'd have to give it

a little rewrite now, wouldn't you?

James: I mean, it's, uh, you, you
got, you got Jake Knapp on speed dial.

It's, uh, it, it, it's, it's,
it's what, 10 years old at least

now, maybe a little bit more.

Jonny: More.

I would, I would imagine
a bit more than 10 years.

Yeah.

Tom: Hmm.

James: So yeah, I, I, I, I would
argue, yeah, it's definitely, I, I,

like, I wouldn't, I wouldn't go back
to like paper prototyping necessarily.

I actually, we, we, we did, sorry,
I'm, I'm off on tangent now, but as

part of this, uh, of this sprint,
we, we did some, uh, some drawing

in, um, in Figma, uh, in FigJam.

And it's been such a long time since
I've done kind of like draw box, uh,

you know, uh, wire framing that I
was like, I, I was going so slow, and

I'm like, we only got eight minutes.

Oh, crap.

So instead, I just, I just
started prompting ChatGPT.

I'm like, right, and this
is what I want to look like.

And then, you know, give it me in this
wire frame, uh, styling, and then just

dropped, dropped these, these images in.

I'm like, then annotated them and
be like, this is what I'm thinking.

Because it was quicker to do that than
me try and put a few boxes together,

which I don't think speaks, uh,
volumes to my, uh, my skill in FigJam.

no, it was, it was, it was, it was
really interesting to see that was the

flow that I ended up going through.

yeah.

Tom: Yeah.

Super interesting.

I remember one of my takeaways from
your talk, James, was the, was the focus

on shrinking the, the learning loop.

And so everything you talked about
there, where the prototyping is in

service of able to move faster and,
and shorten the time to help decide

if a direction is one worth going in.

And I think you've shared some
really interesting things about

influenced content to inform the
product experiences as well that you're

prototyping, which is really cool.

I did have a follow up question
around all of this, which is, um,

had to do that presentation again
now, or that walkthrough again now,

like what, what are, what is the
slide or two that you would add?

Like what are the, what has changed
since you gave that, because that must

have been a couple of months ago now,
and I think a lot of the things you

were showing seemed mature ish or like
they were sort of approaching maturity.

But I'm, I'm curious if there's any
insights or new themes or new that you've

reached now that you have that in place.

James: Yeah, that's a, that's
a good question actually.

I, I think that the core
value is definitely the same.

It's, it is definitely the learning
loop, but I do think that we.

The addition is definitely, there's
a greater depth of collaboration.

Now, I think, uh, and we, I've talked
about it a little bit earlier on, but

there's, there is definitely like this
expectation that these prototypes, whether

it's a, a link shared or a video, they,
there is this expectation that they go

along with either, you know, a feature
that you're building or like I was

talking about in planning, like we, we
are expected as part of planning to share

these prototype of what we're, and you
know, it is, it's understandable, but when

you can't really explain what, uh, you
know, a a, a sort of non-deterministic,

UI is gonna do in a few, few screenshots,
it's really hard to get, get that across.

Especially when you're talking all of
these, uh, strategy plans and proposals,

you know, they go through our, our
senior leadership and our C-suite and,

and you know, it's what we're gonna
spend the next six months working on.

You need to be concrete
on your bets, actually.

You could end up wasting
six months essentially.

So, yeah, it's, it's great that
they've been such a deeper point of

collaboration where there's, yeah,
there's expectation now that they exist.

And I can only imagine that, you know,
if we had this conversation in another

two months, that would a only be deeper.

And the other thing I would really like
to see when I'm starting to see a bit

of is the, the kind of handoff, and
I mean, I would expect it to be less

of a handoff and more of a handshake
experience between design engineering.

we're already, uh, trying to play
around with how could we bring

that react prototype into, uh, so
Figma code layers, for instance.

So, you know, for anyone that
hasn't seen the Config, uh, uh,

demos, but like they've, they've
kind of brought Figma Make to the

canvas, it looks really impressive.

I think that it gives us a lot of ability.

We, what we want to be able to do is,
know, connect all of the infrastructure

that we've built, have it live on
the canvas, and then actually like

a, Linear ticket or some kind of
like, say, like handover handshake

would be, you drop that link in.

And also it could be that it,
uh, you know, tells you what the

components are that have been used.

It could give you a spec.

That's the kind of thing that
I would see as the next step.

But equally the reverse, not just
how do designers make more code?

How do we get engineers and product
managers getting into Figma, getting

into those prototypes and then saying,
well actually we could do it like

this duplicate boom, here's my idea.

'cause I don't see a lot of, a lot
of that I see with products sometimes

that they might grab your frame and
give you a rough idea or, or butcher

it, whatever you wanna call it.

Um, but I do, I do, I I haven't
seen a lot of it from engineering

where they would like design.

Whereas I think if we could start
to bring this like experience where

it is like a connection of code and
design, I would love to come and have

like a, a collab session, a jam with
engineers, with product managers.

I think that'd be great.

So I think the next, you know, we, we
can have another chat in two months.

I don't mind in two months I would love
to be selling you that not only are.

Design utilising this to to, to,
know, shorten the learning loop.

We're using it at a strategy level
to set vision and direction and the

sense expectation that these artefacts
go along with, um, the actual,

uh, features that we're building.

I would love to say they're
not artefacts anymore.

It's not, it's not just a thing that we
look at, it's a thing that we actually

use that we visually collaborate on.

That's the, the next sort
of step, in my opinion.

That is certainly what I would
like to see more of at Monzo.

But at the minute it's been really hard
'cause the, the tools don't really allow

for that level of collaboration yet.

You know, you don't have
multiplayer on, on, on, on Cursor

or Claude Code for instance.

So think that that would be a really
great, uh, piece if we can get it

to work in the way I want it to.

I, I definitely think there's layers
to that, a bit, bit of a amisu.

I think we could quite easily
on a really small interaction.

You, you don't need to go to
the depths of like pulling the

entire, um, design prototype repo.

You could just say, I want to play around
this interaction, and then you might have

a few riffs off of that with other people.

but yeah, I, I think that for the, um,
the sort of depths of where we want

to go, um, with the design prototype,
I really do think it's about just

trying to choose the right path.

So if collaboration is the key element,
keep it light, keep it on the canvas.

If building something that mimics the app
experience, we wanna test it with users.

Do that in the React Native prototype.

I think this, this is what I would like
to sort of see the progression of it be.

Tom: You are, you're bullish on, on Figma
still having an important role to play or,

James: yeah.

sorry.

Yeah.

Um, yes, I am, uh, I went to Config
this year to say, what are you gonna

do to make me open you every day?

That, and like, I'll be
completely honest about that.

I said it, I said it to a lot of people
because I wasn't, I was, I was purely,

and again, I I, I do say this with a
caveat of, I have a very privileged

position where I don't have to open
it to handover with a squad every day.

So I, I do recognise that I still
think it is the predominant way that

teams collaborate, but I was personally
not using it on a daily basis.

And mostly only if I was going to kind
of play around with other people's

work or, or butcher their work, uh, or
be trying to build something, I was,

you know, mostly using it as a context
source for an agent of some sort.

it was, I very much went to Config to
be like, what are you gonna do about it?

And I, and I, and I, honestly, I do
think they popped up with a great answer

because think at the minute, the, the
thing that cannot be matched is you

still have, designer knows how to use
Figma and they know how to use a canvas.

And that is their, like still their
primary sweet spot, their primary tool.

So if you can then their skills
without them needing to leave the tool.

'cause this has been one of the
adoption problems that I think

every team has been facing.

It's like, how do you get designers to
go and open an IDE to open the terminal,

you know, and start, start hacking?

Uh, how do you get 'em to do that?

So the, it is quite clever that the way
they've done it with Figma is just that.

Yeah, let's bring it to the Canvas.

I, I don't think Figma Make was the thing,
I think when they built Figma Make, they

were trying to stave off Lovable when
they've now thought about it properly.

Sorry if anyone from Figma ISS
listening, but when they've, they've

had a minute to think about it.

And actually the thing that they really
need to stave off is not how do we

keep up with these other no code tools.

It's if people are gonna move to
work in an IDE to code first, how

do we, how do we keep up with that?

How do we make it that that can be done
where you, where you work in Figma with

all of Figma other, uh, great features.

So I think the collaboration aspect,
comments, annotation, you know, you can,

you can even use of the Figma motion
piece, like I've been playing around

with, with Figma Motion, bring it into a
component, drop that into a, into a, a,

a prototype it all, it's all connective.

Same with Weave.

Like you've got that, um, the node
based tool, how it kinda connects

things and if you change like the
source, it will update the whole piece.

That's sort of what I
expect now from tool.

Just, it's like, I think your
expectation just expands and you're

sort of like, you know, if I was gonna
encode change a component, it would,

propagate through the entire code base.

So if I change a static component, it
should propagate through the whole thing.

And so that applies to these,
um, to these code layers as well.

And I think Figma Make didn't
really do that very well 'cause

it was so detached from Figma.

I think, you know, the same kind
of problem that they've had with.

And insights and FigJam, et cetera.

It wasn't on the core
product, the core canvas.

So I, so I really do think that they've,
uh, they've come out fighting and

I, it's certainly given me a reason
to open them and play around again.

So I, I, and I do think for those
folks that are maybe less keen to

open Claude Code or Cursor or whatever
you tool of choice is, it's a really

great way to start learning things.

'cause you, you can start to learn some
principles as well, like, you know,

duplicate a frame, it's, it's a branch.

How do I make edits and then
I, you know, I wanna save them.

It's kinda like learning Sketch again,
you know, command S is, well, that's

actually a commit, you know, you've gotta
sort to understand some of these things.

Uh, I have spoke to the Figma team
and one of the things I said was, I

understand the, the desire to, and I
don't mean dumb it down, but to sort

of it more user friendly, friendly.

I think auto layout is
a great example here.

Also, layout is really easy to use, but
it does not help you understand Flexbox.

In fact, I'd argue it makes it
even worse because you're like,

hang on, I want it to be there.

I've said flex-start,
and then it's over here.

And so if they'd, if they'd have
taught, if they'd have found a way

to talk, to teach people the auto
layout, how it actually connected to

Flexbox, even with just like the subtle
label naming, maybe, that would've

gone a long way from designers, you
know, seven years ago starting to use

auto layout to now trying to figure
out what the hell flex start means.

Jonny: Mm.

James: I

Jonny: Although James, it, it
doesn't matter 'cause no one's

gonna read the code anymore, so.

James: I mean, I mean you say that, but
Well, so, so this is, this is the other

thing there that bothers a little bit that
bugs me, which I really think there needs

to be a solution in Figma for, is that,
you know, if I, I, I do go in and I will

make code and it's because sometimes not
necessarily won't work, but sometimes I'm

token conscious and so I don't, I don't
wanna spend money just to move something

from here to there when I know what to do.

Same with like anything, if you want,
you know, you want it to understand,

rounding or you set a specific tech size,
and then I understand they built the, the

kind of, uh, inspect panel in, in Cursor.

But that is clever.

The way they did it was that
it still needs a, a, a pro,

it just gives you a prompt.

It doesn't actually change it.

And it's, it, I say clever because they,
you know, they've, they've cashed in.

But what, what bothers me
is I don't wanna do that.

I don't wanna have to pay for that.

And I, and I do think if there
was a little bit more of a

habitual learning loop between

Jonny: Hmm.

James: doing and the code, it would be
a lot better for, for us as designers of

how we're actually learning these things.

Jonny: Yeah.

James: think that's the, that has
been my feedback on code layers is

like, how can you actually connect
what people are doing to how it

really is working in the background?

Like, how is it working with Git?

'cause you know, you, you take a
designer through GitHub for the first

time and they're like, what's a commit?

What's a branch?

What's a pr I don't, I don't get it.

And then you start trying to like,
describe it as, you know, well, imagine

I copy your file and I put it over
here and I do whatever I want with it.

And it's, you know, you, you
try and go down those roads.

So I, I really hope that they
take the opportunity to, to

teach a little while, while

Jonny: Yeah.

James, I'm pretty sure you brought
Abstract into, uh, Deliveroo

back in the day when that was
the original for Sketch file.

Uh, Sketch files.

Yeah, Sketch files.

Tom: Yeah.

Jonny: What I think, um, let's
move on to the quickfire round

James: Ooh.

Jonny: and James Quickfire
is quickfire for a reason.

James: I have no idea
what you're talking about.

Jonny: Uh, so, uh, my, I, I
have, uh, a quick fire question

and then I'll hand over to Tom.

So, uh, last week we talked quite
a lot about managers and senior,

senior designers burning out and
sort of feeling, feeling like they're

behind on the tools and behind on the
craft, but also, and that that's an

expectation of them all of a sudden.

Um, any tip, tip or two for someone
who is in that position where they're

like, I'm suddenly feeling like I
need to actually be designing again.

I haven't designed for a little while.

Uh, you know, how would
you get back into it?

James: So you're sort of asking like
what's the the opposite to using AI?

Is that the question?

Jonny: I mean, I suppose this
is in a world where AI isn't

gonna just do it for you.

Yeah.

James: Gotcha.

Yeah, I, I mean, I think, I mean, it can
be nice to have a palate cleanse, right?

I think that getting into a project
where AI isn't gonna do it for you,

getting your head in, and that, if that
isn't possible at work, maybe something

in, in your, in your personal work.

I think just getting into something
that's like really design focused.

You really wanna solve a problem
and you don't necessarily wanna

think about how am I gonna build it?

How, what's the end product?

What, like, you just genuinely
get into some craft execution.

Uh, I would call that like
a nice little pallet clean.

And I, I honestly like, I mean, I, I
know I was talking about my holiday

last week before we started recording,
but I did, I did build something, but

it was a massively different thing,
complete pallet cleanse for me.

It wasn't anything to do with work,
it wasn't anything to do with any

of the projects I've got on the go.

And it was just fun and I enjoyed it.

Uh, and I think that for me is, is what
sometimes what you need, and it can stave

off, stave off the burnout, I think.

But equally, I know this is quick fire,
but equally, I think that put so much

pressure on yourselves, like, you know,
just, just there's so much going on.

Try and focus on the things that help
you and that improve your workflow and

improve the things you are good at.

Don't think about every tool that comes
in the market, just, you know, narrow.

Jonny: Brilliant Tom.

Tom: James, when AGI takes over
everything, what will you do

with all of your free time?

James: Oh, uh, well, you know, I actually,
I think we touched on it earlier.

I would probably, still make
things like a hundred percent,

I'd still be building stuff.

But if you, I guess if you assume that
we have this, you know, uh, uh, greater

than human intelligence where anything is
possible, so not just building software,

building hardware, probably start going a
bit bit big on my, my project ambitions.

Like, I don't know, start trying to
build like an Iron Man suit or something,

or, or, or I dunno, a spaceship.

Can I build a spaceship?

Build,

Jonny: So you, you were serious about
this robotics designer thing, eh?

James: Yeah.

I mean, like, but my, this is my point.

If, if, if something more intelligent
than me and more and like, more

impressive than what I can learn on,
on YouTube tells me what to do, I,

I, I'll, I'll start building stuff.

Yeah.

I think that, I dunno, I'd
probably get divorced, but I,

but I'd be building things.

I, but then wife would be like, why
is there a spaceship in the garden?

Well, AGI.

AGI..

Jonny: Related, but have you seen
this website MeatLayer where, um, you

can sign up and it's like a Fiverr
or whatever, but agents can ask

you to do things in the real world.

So if an agent has to complete a task,
like post a letter, then it can pay you to

go to the post office for it completely.

Uh, you could do that.

You could be, yeah, you
could, uh, you could be,

James: meat.

Jonny: yeah, be meat
for the agents anyway.

James: Wow.

Jonny: uh, I've got one more, uh, and
uh, and maybe we'll do, do one more each.

Uh, so what is, it sounds
like you're a Figma bull.

We've had some Figma bears
on, on this podcast as well.

Um, aside from Figma, what is your
favourite product at the moment?

James: Oh, I can tell you my most used
actually, and then probably favourite

as well, but I, I've been using Hex,
uh, it's a dictation app, but I like

it, which it's open source, so I managed
to like, on top of it, build some of

the features that I wanted it to do,
to have, make it a bit more like Wispr

Flow, but without having to pay for it.

Sorry, Wispr Flow.

Um, if you ever want the one that's
a sponsor, I've just killed that now.

So, um, so that's definitely my
most used because, because I, I,

I pretty much don't type anymore.

Um, and I'd say favourite,
am I allowed to say Monzo?

Is that, is that, is
that too much of a plug?

Jonny: I mean, it's, it's a little
bit embarrassing, but you can say it.

James: Oh, no, I know.

I, I think, I think probably
Cursor is, is my, is my favourite.

I, I genuinely, it's the first
thing I open in the morning, so I

would say that would, that gives
you uh, of enough to know there.

Tom: Fascinating.

Um, one very niche specific question about
Monzo since we, I, I previously worked

in Monzo, you are, you're at Monzo now.

I'm curious what the best
AI themed Slack reaction is.

James: Uh, there's some
really good ones actually.

I, I would say it's really the, the,
the, uh, AI image game has really

taken off in the last, like 12 months.

And I think my joining has got
nothing to do with that whatsoever.

Uh, but any,

Tom: have, uh, led that charge.

James: do, I do, I do enjoy it.

Um, but I think anything that
gets created immediately gets

turned into a Slack reaction.

So I, I, there's some
really cracking ones.

Uh, you, we've got our design system
lead, uh, Sam, who I'm sure you, you know

very well, Tom, uh, and he is our sort
of resident apple, uh, expert as well.

And he, he recently had a really big
milestone and uh, uh, at work and I, I

turned him into a, um, a sort of Sam Jobs
character, uh, you know, get good old.

I think I might have done that to
you one, uh, back in the day as

well, Jonny, the old Silicon Valley,

Jonny: I think so.

I think we were both in, yeah, we
were both in this, in that poster

James: The hair, I

Jonny: moment.

James: hair.

Jonny: Yeah.

James: but yeah, so I did that
and, and now there is a, a Sam Jobs

reaction, so anything re remotely
Apple related gets, gets the Sam Jobs.

And there's, there's a, there's a
few other really good ones, but yeah,

I've got an entire folder, like just
like AI, slop images, basically.

Tom: What and what, what your,
what is your tool stack of choice

when generating such memes?

James: That's a good question actually.

I, I found a lot of joy with
ChatGPT actually recently I tried,

I had a, a whole Gemini period
'cause I thought Nano Banana was

pretty good when it first started.

But I tend to find the, the drift on
Nano Banana now is like really wide.

You can sometimes get really
good output and sometimes awful.

Uh, whereas I think it's a little tighter
on ChatGPT you might sometimes need a

couple of revisions, but I also think
the UX on ChatGPT image generation

has got a lot better because you can
describe more specific, uh, edits.

You can obviously use circle and annotate
and things like that and you can, if

you, if it wildly misses, which it really
frequently does, if it wildly misses

the reference photo for instance, you
can drop it back in and say, no, this,

and there's, yeah, I dunno why, but
whenever I do a picture of of, of me and

others, everyone else comes out great.

And I, I, I look like
some kind of burnt cake.

Uh, so it's, I'm not entirely
sure what's going on there.

I think it's how to get me, they're
trying to reduce my token, uh, usage.

Jonny: Well, if we clip one thing from
this podcast, James, that'll be it.

James: The The burnt cake.

Oh, surely not.

Jonny: No,

Tom: Amazing.

Jonny: you said some good stuff.

I'm sure we'll find something to
clip that eclipses the burnt cake.

Um, right.

Tom: Let's

Jonny: think, uh, I think,
I think that's wrap.

Uh, James, thank you so much for
joining us and, um, really excited

to see what you continue to build
at Monzo and keep pushing AI

forward and all that kind of stuff.

London is a fintech capital and
we've had two fintechs so far.

We'll probably continue.

So, um, uh, great work there and,
uh, thanks everyone for tuning

in and we'll see you next time.

James: Awesome.

Thanks both.

Really appreciate it.

Bye.