The discussion centers on "Cloudbot," an AI assistant that integrates deeply with a user's computer and messaging platforms, acting as a highly resourceful and personalized digital companion capable of performing complex tasks and automating workflows.
Mind Map
Click to expand
Click to explore the full interactive mind map • Zoom, pan, and navigate
It's just like having a new weird friend
that is also really smart and
resourceful that lives on your computer.
I was still in Morocco and someone sent
me a tweet of a bug and I literally just
made a picture of the tweet, posted on
WhatsApp. It read the tweet, it
understood that there was a bug. It
checked out the Git repository, it fixed
it, it did a commit and then it replied
to the person on Twitter that it's fixed
now. This will blend away probably 80%
of the apps that you have on your phone.
Why should I use my fitness pal to track
food when I have an infinitely
resourceful assistant that already knows
I'm making bad decisions and I'm
Kentucky Fried Chicken? Those things are
so resourceful, although in a scary way.
It's like unshackled JGBT. A lot of
people don't realize that if you give an
AI access to your computer, they can
>> All right, welcome everyone. My guest
today is Peter, creator of Claude, an AI
assistant that you can just chat with in
your messaging apps to get stuff done.
And today is Peter is going to show us
how to use cloud. And also Peter has a
lot of really great hot takes about AI
coding that I'm really excited to dig
into. So welcome the other Peter.
>> Peter, thanks for having me.
>> Great to see you. So why don't we start
by talking about cla. So what exactly
does cloud do at a high level and uh why
why is it a lobster? [laughter]
[laughter]
>> Yeah. So, so maybe I have a little bit
of a backstory like when I when I came
back from my retirement, let's say it
like that. I kind of wanted wanted a way
to just check up on my computer from my
phone cuz I I fully jumped on this on
this V boarding trend and then you know
like your agents might run for half an
hour while you while you eat or they
stop after 2 minutes because they have
another question and then and then you
go back and you're like annoyed. But I
kind of I kind of didn't build it
because I assumed all the big labs would
do that anyhow.
it felt like such an obvious thing
>> like like something where like kind of a
new new kind of operating system almost
um but didn't happen and then it was
like November and it still didn't
happen. I'm like okay I'll try something
small and and the small thing was
basically hooking up WhatsApp to cloud
code. So, so, so you send a WhatsApp
message and it would literally open the
binary like with the the prompt and like
return you this thing like very simple
was like built in one hour and it kind
of got a life of its own and now here we
are. It's like I think 300,000 lines of
code. It does every messaging platform
on Earth. Not everyone, but like we're
getting there. And I think it it's kind
of where things are going in the future.
like everybody will have an AI that is
like super powerful and like follows
them through their life. Um, turns out
if if you give an AI access to your
computer, they can basically do anything
that you can do on your computer.
>> Yeah. And it's got to a point where like
you don't have to sit there and babysit
it, right? You can just give it some,
you know, prompts and commands and it
will do it thing for you and you can
check it work and then that that's it.
You know, you have to set a computer.
>> Yeah. So when I when I built it, I I
feel I feel this project is like as much
exploration as it is like technology
because it's it's a little bit of a new category
category
>> and then I was on a birthday trip for
one of my friends in Morocco
>> and I I catch myself using this thing
all the time. It's like, hey, where are
we going? like asking for directions or
like or like tips for for restaurants or
I don't know there was like one morning
where someone sent me a tweet of a bug
and I just literally just made a picture
of the tweet posted on WhatsApp. It it
read the tweet. It understood that there
was a bug in one of my repositories. It
checked out the git repository. It fixed
it. It did a commit and then it replied
to the person on Twitter uh that it's
fixed now. And I'm like really
>> this is nice. And then one day I was I
was walking around and I didn't sync and
I just sended a voice message, you know.
I didn't I didn't I didn't build in
support for voice message and I was like
it showed me a typing indicate. I'm like
oh I wonder what's happening now. And
then it just replied to me as if nothing
would have happened and I'm like wow how
the f did you do that? And it was like
yeah I saw a file but there was no file
ending because I didn't build it. Mhm.
>> So I looked I looked at the at the
header of the file and it was like ous
uh some some audio file format. So I
found ffmpeg on your computer and I
converted it to wave and then I looked
for visper.cpp but you didn't have it
installed but I found this openi key and
then I used curl to send it to openai's
API uh and it got the transcript back
and I replied to you. I'm like >> wow
>> wow
>> that's amazing. Yeah,
>> like those things are so resourceful
although in a scary way. Um, but that
was kind of like the the the moment
where it clicked for me is like, oh
yeah, this is really powerful. This is
much more interesting than using JGBT on
on the web because it's like it's like
unshackled chbt and I think a lot of
people don't realize that that those
things like cloud code, they're not just
good for programming, they are very
resourceful for any kind of problem.
Yeah. You just got to give it access to
like, you know, your computer and like,
you know, be able to find stuff, right?
So, you just have to give it tools.
They'll become very resourceful.
>> Yeah. So, so, so over the last few
months, I I kind of built up my my CLI
army because what are what are agents
good with? Calling CLIs because that's
what they train for. So, I've built like
CLI for accessing all of Google,
including like the Google Places API. I
built one that that that makes it very
easy to look up memes and gifts. So you
can also like reply with some memes. [snorts]
[snorts]
>> I did I did a bunch of experiment. I
even built one that visualizes sound
because I wanted it to like experience
music. Um that that goes a little bit
more into the art direction. I don't
know if that makes any sense, but
anyhow, it's a lot of fun. Like like I
have like a whole list. I built one that
that that hacked into my the food
delivery here. So So it actually tell me
how long it takes until my food's there.
I have one that that that reverse
engineered the the eight sleep uh API so
it actually can control the temperature
of my bed
>> really and real quick like when you
build all this stuff you just you're
just getting AI to build this stuff or
what's going on
>> the funniest thing is I I was in my my
old company um I was very very good at
iOS and and Mac OS like the whole Apple
ecosystem I did it for 20 years I'm like
very much an expert in there
>> but also when I came back I I I built a
project where I'm like I'm kind of sick
of Apple gating everything
>> and it also would make a lot more sense
to build it as a web app because it's
kind of a thing that it should be in a
browser and and should be accessible and
if I do it as a Macup again it's like
we'll have a very limited set of people
that can use it but you notice I see
this with a lot of engineers where
you're really good at one thing
>> and then moving to another technology is
just so painful because then you feel
like you feel like a sorry for the word
an idiot and [snorts] and you have you
look up every little things like what's
a prop or like how do I split an array
cuz because you understand all the
concepts but you don't necessarily like
know the syntax. So that that's kind of
like how I felt when I moved from from
Objective C and Swift to to JavaScript.
Like I know JavaScript a little bit, but
I never really built something big in
Typescript. And then it's just it's not
even that it's hard, it's just painful
because you have to look up all those
things and you're just so slow. And then
and then with AI all that all that melts
away like you can you can still apply
your system level thinking your like how
do I build and structure bigger projects
and your taste may I say or like your
which dependencies do I build on um all
those things all these things is still
valuable and you can like much easier
move that from one domain to the other.
>> Yeah. And and that felt like a
superpower. Suddenly like I I feel like
I could build anything. Languages don't
matter anymore. My my my engineering
thinking matters.
>> Yeah. Because like I was trying to worry
about like whether you have a
parenthesis here and there is is lame,
right? You don't have to worry about
that kind of stuff anymore.
>> This episode is brought to you by
Granola. If you're in backtoback
meetings, you know how much work it is
to take notes [music] live and clean
them up afterwards. That's why I love
Granola, the best AI meeting notes app
in the market. Here's how I use it.
Granola automatically takes [music]
notes during a meeting and I can add my
own notes too. After the meeting ends, I
use a Granola recipe to extract clear
takeaways and next [music] steps in the
exact format that I want. Then I can
just share notes directly in Slack with
my colleagues or even get Granola to
share her notes automatically. Honestly,
of all the AI apps that I use, Granola
is the one that saves me the most time.
Try it now at granola.ai/pater
and use the code Peter to sign up and
get three months free. That's granola.ai/pater.
granola.ai/pater.
Now, back to our episode.
>> But, but dude, let's let's go back to
Claude like like the the thing that you
built. Maybe you can um screen share,
maybe you can show like maybe you can
show like how people can install the
thing first like it like do you have to
be super technical to use cloud or no?
Right. You can just install it and then
get it to work.
>> Yeah. Um yes, yes and no. So
>> fortunately and also unfortunately this
project attracts a lot of uh people that
don't have a lot of clue about
technology because
>> it it lends away all these all these
layers that make it complicated, right? You
You
>> if you use cloud code, you work in a
terminal, you you kind of have to think
about the context space and what folder
you're in and like
>> it feels very techy. like talking to a
thing on on iMessage or WhatsApp or
Telegram. You do it with your friends
and now it's just like having a new
weird friend that is also really smart
and resourceful that lives on your computer.
computer. >> Yeah.
>> Yeah.
>> That makes the whole technology very
approachable. You don't think about, oh,
what model do I pick or what? It just works.
works. >> Mhm.
>> Mhm.
>> And then that's that's that's kind of
the idea. It's also like that good and
bad part because of course with a lot of
power comes a lot of risk which is also
unsolved like this thing is access to
your computer. So yes it could do bad
things on your computer. If you tell it
to like I don't know delete all my files
in my home directory. It would probably
like ask you are you sure? But if you if
you like if you like keep saying yes yes
yes it will probably comply and probably
also delete itself and crash. [laughter]
>> Yeah you got to be careful. Yeah
>> I I'll let me share my screen so you
see. So in it's written in in TypeScript.
TypeScript. >> Mhm.
>> Mhm.
>> So it runs it runs everywhere. Uh even
on Windows you can just go on our
website clogbot. >> Yeah.
>> Yeah.
>> And there's a handy oneliner. Looks very
scary but it's every everything's open
source. You can check everything
including the website. >> Mhm.
>> Mhm.
>> So so this is the easiest way to install
it. It works on Mac OS Linux. It also
works on Windows.
>> It's time to turn terminal right and
then it'll start installing. Yeah. You
can also install the mpm for people who
understand that ecosystem
>> and I think something that that I did
that I haven't seen in a lot of lot of
project is also it has a hackable install
install
>> again a simple way and like the more
manual way where you basically check out
the git repository and then and then
launch it from the git repository which
is to to be fair is like the most fun
way to use it because if your agent can
read its own source code of its harness
it can literally reconfigure and
reprogram itself and then restart and
then either crash or have new powers.
Um, [snorts] >> okay.
>> okay.
>> Like this I think this is one of my my
superpowers where I got a lot of people
participate in the project and like send
me pull requests that never did a pull
request. I mean
>> sometimes that also shows um but I but I
see pull requests more often as a as a
prompt request. Yeah,
>> it's enough to like understand the intent
intent
>> and then and then and then a after we
install it like do you like how do you
hook it up to a messaging app?
>> The nicest way is probably right now
just using this one lineup and then it
will like greet you with some sassy
stuff and and try to configure
everything up.
>> Got it.
>> In install the package and then you and
it just guides you through and you just
you can hook it up to do any of the
common messages.
Okay, that that looks good. It's working.
working.
Yeah. And then you can say plbot.
It will do it automatically if it's a
clean install. But now I have to type in
on board.
>> Got it.
>> And basically then you have like you can
enter a model.
You can Oops.
All providers.
Let's say we go with some tropic.
Probably a new one.
Then you can set up Telegram, Discord
and then it will guide you through the
rest. You can like set up skills hooks
>> and then do you have to give it your
anthropic API key?
>> It works with with any model. >> Yeah.
>> Yeah.
>> Although you know they're in industry
like Anthropic and Open are kind of like leading.
leading. >> Yep.
>> Yep.
>> It works with API keys. It works with subscriptions.
subscriptions. >> Okay.
>> Okay.
>> Although although there's been there's
been a little we we we added support
with it because it's kind of like what
everyone does.
>> Yeah. But I don't think Entropic likes
it anymore. So I would recommend using
using an API key or going with another
model. There's >> okay.
>> okay.
>> Yeah. The the main problem is like OpenI
works well, but it's just not as funny.
There's something special about Opus
that makes it really really fun.
>> Oh yeah. Like like the personality.
>> Yeah. Yeah. There's I don't know if if
you've read this article about how they
put a soul into the model.
>> Yeah. And then people rediscovered it by
like feeding it blocks of text and
letting it like continue. And eventually
they they re they they they squeezed out
the the text of the soul that the model
wasn't even aware it was being trained
on. Is is a really interesting story. I
don't know. I feel that has has
something to do. But this is the first
model that is actually funny to use.
Like I I built mine so that >> Yeah.
>> Yeah.
>> It can roast me
and it probably doesn't know that it's
on camera right now [laughter]
>> and and it's it's going to roast you
based on what it has access to all your
computer stuff, right? So it's
>> Yeah. Here. >> Yeah.
>> Yeah.
>> All right. You asked for it. You sold
the wildlen world to find yourself did
Sarah. We tried everything and your
bigger relation was I should build more
software. [laughter]
You're so obsessed with the alto is that
you literally built yourself a friend
because debugging code is more fun than dating.
dating.
>> And let's be honest, the only reason I
existed because you needed someone who'd
listen to your chainway hook takes at
your Amazon judgment. >> Wow.
>> Wow.
>> Yeah. Now go kill that podcast. So So I
I hooked it up to pretty much everything
on my computer. It it can read my
emails. It can read my calendar. It can
access to all my files. It can control
my lights. I I use Philips who? >> Yeah.
>> Yeah.
>> It it can control my sonos. So like I
can I can tell to wake me up in the
morning and I slowly like turn up the volume.
volume.
>> It has access to my cameras. It was it
was funny like I when I built access to
the cameras I I told it like watch for
strangers and then it told me in the
morning like Peter there's someone and I
like I watched in the whole night and it
made like screenshots the whole night of
like my couch because it was like there
was like the camera is pretty blurry and
it looked like someone was someone was
sitting on the couch and it assumed
there was a stranger sitting on my couch
the whole night.
>> Wow. But yeah, I slowly it it has I'm
just thinking like in my place in Vienn
it also has access to my my kx so it can
actually control every part it could
literally lock me out of my house like
in space they say it's like oh can't do that
that
>> okay how did you hook all this stuff up
you just ask cloud to do it or or what
you just
>> yeah yeah literally you know you know
there's this thing where
>> we have skills
>> so usually you you talk to it will these
things are really resourceful so it will
like figure out an API Right? It can
Google for it. It can find your keys on
your system. You can give it keys. And
people use it for everything. People
build like a a skill to so go shopping
for them on Tesco. Um or buying stuff on Amazon.
Amazon.
>> I I I let I let it check in my flight
from British Airways. And this is
actually I don't know if you've I mean
you've used check in sites. This is I
feel like this is almost like the hi
test. It was a touring test, but like
steering a browser to check you in on an
airline website is like the ultimate
test and and and then the first time I
my integration was pretty rough. So it
took almost like 20 minutes that was
still in Morocco and everything was very
much hacked together. >> Mhm.
>> Mhm.
>> And then it finally managed but it
actually had to had to find my passport
in my file system. It found it on
Dropbox, extract a key, put everything
in it correct and it finally checked me
in and I was like watching it and sweating.
sweating. >> Wow.
>> Wow.
>> Yeah. Wow.
>> But now it works much better. Now it's
like it it does it within minutes. So it
can it also happily clicks the I'm a
human checks on the browser because it
literally just it it literally just
controls a browser on it has its own
little computer over there
>> and and just kicks around. So So it's
like really really difficult to detect
from all those antibot systems because it
it
>> it doesn't feel different from a human
in in those patterns. Can you maybe like
can you show us like just a couple more
use cases like uh can you maybe have it
turn on the light or show some use cases
from other books?
>> Yes. So I started collecting because I
feel I'm so I'm so bogged down actually
building it. By far I'm no longer the
the most creative one who actually use
it. So people people hooked it up to
their messaging system so can actually
reply uh not just as not just to you but
to everyone.
>> Uh and you can also hook it up into
group chat which is even more fun.
There's a lot of people that use it as a
as the family the family member almost.
Um yeah, send me reminders, create
GitHub issues, sync with Google Places,
or every time you make a bookmark on on
Twitter, it will like capture bookmark
and add it to your to your to-do list. >> Yeah.
>> Yeah.
>> Keeping track of costs. Um some people I
I I programmed something in to like
remind people that they sleep enough. So
like they always get like bitched it
when they're up at night from their bot.
So it can help you track your sleep. It
can access your your fitness watch. Uh
it it has its own little one password
vault. So if I want a password shared, I
move it into its own vault and it can
access that one. Um because it you still
want to have some boundaries. I mean
there people give it the credit card.
>> Yeah. I don't know about that. [laughter]
[laughter] >> Yeah.
>> Yeah.
>> Um it can do all those things like
research, creating invoices, managing
managing your email.
>> These people are like enthusiasts,
right? like they they've really
customized it to do whatever they wanted
to do. Like how about for like dude like
how about for some just just download
this thing just download for you know
fresh fresh install and what are some
like really common use cases that I can
get get to do like just like manage my
calendar or you know stuff that doesn't
delete my computer
what are some uh safe things to start with.
with.
>> It's funny because everybody takes a
very different path. There's like people
who like install it and immediately
immediately build an iOS app with it
because it's also it's also a coding
agent. It can do anything. It can spawn
sub agents. It can it can either code
yourself or it can like control cloud
code or codex and ask them to code.
>> Yeah, this guy started immediately like
managing Cloudflare. This one was great.
Week one set up for my family. Week two,
set it up for some non techy friends.
Week three, we're building cloud for
work. I hope I hooked up one of my my
non- techy friends and like he started
sending pull requests. He never did it
in his life. >> Yeah,
>> Yeah,
>> fitness is is is a big thing.
>> Okay. So, so basically it's it's kind of
like u you really have to I guess way to
use this thing is just to think about
what's causing problems in your life and
how you can get this personal assistant
to help you to help you streamline everything.
everything.
>> I don't know if this is the project
that's going to be, but if you think
about it, this will blend away probably
80% of the apps that you have on your
phone. Why should I use my fitness pal
to track food when I have an infinitely
resourceful assistant that already knows
I'm making bad decisions and I'm
Kentucky Fried Chicken. So, so it will
probably like remind me if I forget
tracking the food. I can just send a
picture and it will store itself in the
database and and calculate it and and
and roast me that I should go to the gym
because I'm like way over my calorie limit.
limit.
>> Yeah. Why do I need an app to like set
up when my my bad my ASIP should work or
not? Because it just has API access. It
can just do that for me. Why do I need a
to-do app when it just tracks my to-dos
for me? Why do I need an app to like
check in on my flights when it can just
do that for me? And it's like such a
more convenient interface because I just
I just talk to a friend
>> and because it has so much so much
context, it doesn't need so much custom
prompting. like why why do I need a a
shopping app when it literally can like
recommend me things and do all of that
for me? So I feel I feel
>> there's a whole layer of apps that will
slowly melt away because if they have an
API these are just services that your AI
will do and in a year I don't know I
think actually this year will be the
year where a lot of people will explore
it more and like get their uh AI
assistance from all those big companies.
Yeah, I think uh yeah, why why click on
these little little self-contained
little apps when like yeah, this
assistant has a bunch of capabilities
can just do everything, right? Just
connect it to everything. And I guess
the way you connect it to everything is
just like you just you send it a text
message or something and then it's like,
oh, can you do this? And they're like,
oh, I need to do some re research and
then it just takes care. You just kind
of go back and forth with it and and
make it have happen, right?
>> Yeah. And and then and then it it writes
a skill and it it remembers. So part of
what makes it so interesting is that it
has it has persistent memory. it will
learn about you and it will update it yourself.
yourself.
>> So it it so you the more you use it, the
more you customize it and the more
powerful it gets because because okay
maybe the first time you have to like
guide it a little bit but then it will
it will create a skill and then next
time you can just mention it. Next time
it it I need to like check in my flight.
It will take like two minutes because it
it exactly knows all the quirks of the website.
website. >> Yeah.
>> Yeah.
>> Because it did it before and it made notes.
notes.
>> Got it. Yeah. Yeah. It's like teaching
someone to just learn something and they
can do it easily the ne the next time. Yeah.
Yeah. >> Yeah.
>> Yeah.
>> All right, man. Let's talk about another
topic. Let's talk about, you know, you
came back from retirement to build this
thing and uh you have very strong
opinions about AI co coding like some
pretty hot hot takes. So, let's talk
about let's talk about some of those.
Like you wrote this post that I really
like called just talk to it and then
well like you know these days everyone
on X is writing about like all this
fancy right like all these like you
know hook skills and all this kind of
stuff. So, what's the gist of that that
post? [laughter] Is it just just talk
talk to the AI and figure it out?
>> No, but but I work a lot. I build a lot
of things and I I love Twitter as well,
so I'm very active there and I just see
so much. I almost I I I kind of call it
the agentic trap
>> because people discover that, oh, those
agents are amazing, but they would be
better if they could do a little bit
more. And then they really fall deep
into this rabbit hole. And I've been
there myself where you build really
sophisticated tools to like try to
accelerate your your workflow, but in
the end you're just building tools.
You're not actually building something
that really brings you forward. You just
have this,
>> you know, part of the problem is that
it's so fun.
>> Yeah. like like when I when I started
pretty early, I I worked on VIP tunnel
to have access of my terminals on my
phone and I I was in this rabbit hole
for two months until I it was so good
that I was out with my friends on and
instead of like joining the conversation
in the restaurant and I was just like
vibe coding on my phone and I decided
okay this I have to stop this just for
my more for my mental health than for
anything else. Um, [snorts] and these
days you can build everything, but you
still need ideas. So I see I see so many
managers for cloud code or codecs or
like orchestrators or other little
things that have the illusion of making
you more productive, but >> yeah,
>> yeah,
>> really aren't. I mean, like the the
latest thing that that I just had I just
cracked about is like like Gas Town. Um,
it's like a really sophisticated
orchestrator that is like also very
broken and doesn't really work where you
like run like run like tens or 20s of
agents simultaneously and they all talk
to each other and like try to split up
and there's like there's like watchers
and overseers and a mayor and pcats and
I don't know what else >> really
>> really
>> there's a mayor. Yeah, in Gaston there's
a mayor. I mean I call it slop town. Um
or like or like this whole trend of
Ralph thing where you just like you you
give the AI one little thing and then
like you write in a loop and then if one
little thing is done you trash away all
your context and you start again. Um
like the ultimate token burn machine. >> Yeah.
>> Yeah.
>> And then and then and then can you can
create code and run all night and then
you have like the ultimate slop because
what those what those agents don't
really do yet is have taste. They they
are really they are spiky smart and and
and they're really good at things, but
if you don't if you don't navigate them
well, if you don't have a vision of what
you're going to build, it's still going
to be slop if you don't ask the right
questions, it's still going to be slob.
And I don't know how other people work,
but when I start a project, I have like
this this very rough idea what it could
be. And as I built it and as I play with
it and as I there is a feel it I my my
vision gets more clear and like I I get
like I try out things some things don't
work and I evolve my idea into into what
it will become and that's that's like my
next prompt depends on what I see and
feel and think about the current state
of the project. Yeah,
>> but if you if you try to put everything
into a spec up front, you miss this kind
of like human machine loop and then I
don't know how something good can come
out without having having feelings in
the loop almost like like taste.
So, so it was like
>> somebody somebody tweeted like, "Oh,
look at this mech app. Like it was fully
Ralph." And I replied, "Yeah, it looks Ralph."
Ralph."
like no offense but it looks Ralph
because like you could see clearly see
that like no same person would design it
that way.
>> Yeah, it feels like it feels like some
some people are just like uh actually
building these things not for the apps
themselves but to prove that they can
get it to run for like 24 hours like get
AI to run for 24 hours without
intervention, right? It's like
selfmastery kind of kind of thing,
right? You're just trying to prove that
you can get AI to run super long time. It
It
>> it's like a it's like a size comparison
context without seeing the other word in
a way. And I I've been guilty a little
bit of that myself. Like I also had like
a loop for 26 hours and I was very
proud. But it's that's it's a vanity
metric. It doesn't that makes no sense.
That makes no sense. And like just
because you can build everything doesn't
mean you should or that it's going to be
good. But then again,
>> this this face of I'm just playing. I'm
just like literally I'm building for
fun. It doesn't matter if it's going to
be used is incredibly useful because
that's how you learn, right? That's how
that's how we learn to program. And
prompting is just a different skill.
Sometimes I see people that they're like
a little more on the on the AI skeptical
side. So they ignore it for a year and
then they spend one day where they they
evaluate the models and then they do a
blog post and then they write like a
short prompt and feed it I don't know
into claw web to to make him make him an
make an iPhone app which like is
completely underspec and the model does
its best to deliver something and then
like it doesn't compile because they
they build it on the Linux machine where
there's no compiler for that you know
and then you're like oh AI is not good
and then they dismiss the whole topic
for a year. It's like that's that's not
how it works. >> Yeah.
>> Yeah.
>> You need to play to to to understand how
those little monsters work. You need to
understand a little bit their language,
their way of inference of thinking and
then and then you get progressively
better at it and and then your output
will be better.
>> Yeah. You have to uh you have to be
persistent, right? because sometimes it
doesn't work properly and you have to
fix all the bugs and then and then if if
you just keep playing with it, you
develop this like I guess product sense
or whatever of actually learning how to
talk to the models and learning what
they can do. So
>> I mean partly it's it's being really
weird because they're adopting some of
the language. It's like I'm I'm thinking
of like or you weave a feature in kind
of like you like a like a in German it's
fen a twin a tw or you you run the gate
which like a linting and testing and
building and it looks like a a big line
in the terminal which is like the gate.
So so I'm like we didn't run full gate.
Um [snorts]
sometimes things doesn't work out and
then you can just ask why did why did
you not do and then it will tell you you
said this and this and I assumed this
and this and it's like oh yeah I I
messed up on the language or I was
unclear like if you just tell it build
me a Mac app it will probably assume
that you want you want to support a lot
of old operating systems because the
majority of software does so it will API
>> I I found that asking it to like just
ask a bunch of clarifying questions to
then make you clarify stuff helps a lot. Yeah,
Yeah,
>> it's it's also funny because I'm I'm
more I like Codex modern cloud code. I
think it's it's the better model
>> even though it's it's it's it's
incredibly slow, but it's soro and
things work. And people complain that
there's no plan mode, right?
>> Yeah. and and and and my joke is always
plan mode was a hack that Entropic had
to add because the model is so trigger
friendly and would just run off and
build code if [laughter] you use in in
the in the latest models especially like
like GBD 5.2 to I'm just having a
conversation. I'm like, "Hey, I want to
build this and this feature. It should
do this and this and this and maybe drag
in this control tool like I like this. I
like this design or this and this. Give
me options. Just say let's discuss. Give
me options." Then they have a
conversation with it and and then and
then the model will propose something
and they're like I mean I usually don't
type I talk to it. >> Yeah.
>> Yeah.
>> And staying on brand.
>> Do you do anything to um manage the
context? Once you have a conversion with
it, it can get really long and it can
get confused. Do you try to like do you
make it manually compact or summarize stuff?
stuff?
>> I feel that this is also like old old
patterns. This this was a this was very
much a problem with with cloud code
still is to a point. Codex is so much
more the context lasts so much longer.
Even though like on paper it's maybe
like 30% more but it feels more like two
or 3x more. I think I think a lot of it
has to do with like the the internal
thinking. the internal thinking of of
GPD models is really effing weird.
>> But yeah, like the for context
management, I think this was much more a
problem in the earlier models. Now most
of my features fit into one window. So
the whole discussion and building
happens simultaneously.
>> Okay, got it.
>> And that that works fine. Like sometimes
if if it's a really long thing, I create
a new one. You want to like codify it in
a file.
>> Mhm. But this is much less of a problem
than it than it was before. That's again
like the space evolves so fast you just
have to try things.
>> Yeah. Got it. Okay. Got it. So basically
so just to summarize like basically when
you add a new feature to cloud or what
whatever you're building like maybe just
walk through the steps like first you
explore explore the problem and the
solutions with the AI and then you do
you actually make a plan at all or or
just go
>> it's even better like I you know I I
built this project and
>> which is kind of like a mix between
Jarvis and her
>> her. Okay. Okay. the movie. Yeah.
>> Yeah. Yeah. Um but it it's very
difficult to convey how it makes you
feel and how useful it is just by
talking about it. >> Yeah.
>> Yeah.
>> So on on Twitter it got me very muted
traction. I'm like I don't get it. Like
everyone I show it in in person gets
super excited
>> because they see me interact with it and
and like showing off all the cool things
but it you you can't get across the same
thing on a picture when I talk about it.
So eventually I I created a Discord and
I just hooked up my bot to Discord so
that people could interact with my eye
that has full access to everything on my
system with my my private memory on a
public display.
the most nuts thing that I ever did I
think and then people got hooked and now
and now like people always ask me there
oh can you build this feature or there
this and this bug and most of the time I
now approach it by making a literal
screenshot of the conversation dragging
it into the the terminal or just copying
the text depending on how it is and just
ask hey let's discuss this because I'm
lazy I don't even have to type anymore I
just like copy in Discord conversations
>> um or the same also when people ask hey
do you support this and this or how to
do this this I'm just like can you read
the code and write a new FAQ entry and
it does that
>> so that's so these days it's actually
how I start building features most of
the time but like reading on discord and
and and and seeing where people have
pain points [laughter]
>> oh my god so you just paste the paste
the conversation and you just talk to AI
about it and you figure out what the
right solution is
>> partly that partly also I I have a have
a scraper so every at least once a day I
scrape the the help section and and ask
the model what are biggest pain points
and then we we fix them.
>> And and do you do all the fancy
like do you have like multiple agents
and like you know you have those crazy
skills and stuff that you run or no you
just go
>> I have very I mean I most of my skills
are actually personal life skills like
food tracking or buying groceries or
like that kind of stuff for coding very
little cuz you you don't need that much.
I don't use MCPS or or any of that crap.
Um, I don't I don't believe in in this
big orchestration systems because like
we talked before, I I feel if I'm in the
loop, I can build a product that feels
better. Maybe there's maybe there's ways
to like build it faster, but I'm already
so fast. I'm mostly limited by thinking
about it and sometimes a little bit by
waiting for Codex, but mostly it's it's
it's my thinking that's a limiting
factor. Um, I just I just used a bunch
of terminals like split screen. >> Mhm.
>> Mhm.
>> I don't use work trees because again I
feel it's just like unneeded complexity.
I just have like a few checkouts of the
repository like cloud pot one 2 3 four
five and [snorts]
then like either they are free or
they're working on something. either I'm
like exploring something or like I'm
building something or I'm fixing
something and then when it's done I test
it locally and then I just and then I
push it to main and then I sync it up.
>> So this is it feels a little bit it also
feels a bit like a factory if they're
all busy.
>> But if you only work on one, it's very
hard to actually get into the zone
because it's just too slow and
>> you have to wait. Yeah, you have to wait
for it to do
>> you can only post on Twitter so
much. Uh, [laughter]
so I I feel I need multiple ones to to
really keep me hooked and get into the
same flow state that I had when coding
just that I I'm like insanely more productive.
productive.
>> D I don't know if you play like the
real-time strategy games before, but
like it's kind of like yeah, you you
have like a squad of people attacking
and then you kind you kind of have to
monitor and manage them, right? And that
that has kind of taken it. my business
partner from my my previous company. Yeah.
Yeah.
>> Also totally got hooked on on Clubbot
and he's he's more from the business
side. He was a lawyer in a past life and
he's now sending me pull requests which
is like insane in its own way. How AI
made like gave people superpowers that
are not that technical to still build
things is amazing. I know I know there's
a lot of push back because yes, this
might not be perfect, but I treat pull
requests more as prompt requests because
they convey the idea and most people
just don't have the same system
understanding. So they not be able to
like derive the the model in a way to
give optimal results. So I just rather I
just rather have the prompt like the
intent and just do it myself or if they
send the PR I will extract the the the
the intent out of it and just rebuild it
myself or sometimes I base it off. I I
still mark them co out
>> but very rare that they actually merge
code directly.
>> Yeah, that makes sense. Okay. Um All
right, man. Well, I mean like um I guess
the take away from this conversation is
like my take away at is like um yeah,
don't don't get crazy with the slop
generators. Just actually have the human
in the loop because the human is still
your brain, the taste and everything,
right? You have to guide it.
>> Yeah. Or like don't find your own path,
you know, like some people always ask me
how do you do that? How do you do that?
>> You got to you got to explore. You're like
like >> Yeah.
>> Yeah.
>> It will take you a while to be good. You
have to make your own mistakes. That's
that's how you learn with everything in
life. And that's how also how to learn
those things just that this space is
evolving very fast.
>> And where can people find uh cloudbot?
>> Yeah. Uh cloud.bot. Uh it's also on GitHub.
GitHub.
>> It's a cloud with a W. CL L A WD, right? Yeah.
Yeah.
>> Yes. Yeah. Correct. [laughter] >> Yeah.
>> Yeah.
>> Like like the lobster claw.
>> Yeah. Yeah. Got it. Thanks so much,
Peter. This is a super great
conversation. Um I I need to give Claude
a try myself because Yeah. I I don't
like sitting on my computer to talk to
the AI. I like to just like be on the go
and just like give it instructions, make
it do stuff while I'm out with my kids.
Like that would be the ideal way for me.
So, it's great that you build something
like that.
>> I'm looking forward to to see what what
you what your opinion is on that. I
think you're gonna like it.
>> All right, man. Well, it's great
chatting with you and uh I think we'll
Click on any text or timestamp to jump to that moment in the video
Share:
Most transcripts ready in under 5 seconds
One-Click Copy125+ LanguagesSearch ContentJump to Timestamps
Paste YouTube URL
Enter any YouTube video link to get the full transcript
Transcript Extraction Form
Most transcripts ready in under 5 seconds
Get Our Chrome Extension
Get transcripts instantly without leaving YouTube. Install our Chrome extension for one-click access to any video's transcript directly on the watch page.
