November 18, 2025 • 49 mins
What if AI could make your work more creative instead of more crowded? We sit down with Scott Werner to unpack a practical path for Ruby developers who want the leverage of AI without sacrificing taste, clarity, or joy. From agentic coding with Claude Code to context-rich tools like Tidewave, we walk through how better inputs—logs, DOM access, database state—turn generic suggestions into usable plans that reduce cognitive load and speed up real problem solving.
Scott shares the origin story of Artificial Ruby, a New York meetup that started as a casual happy hour and became a monthly mini conference. That community energy matters: many devs began their careers remotely and missed the spark of live conversations. By focusing on play and curiosity, the group channels the early Ruby vibe—ship small experiments, trade sharp feedback, and rediscover the fun of making software together. That ethos powers Scott’s projects: Monkey’s Paw, a prompt-based web framework that leans into expressive generation, and Latent Library, a hallucinatory book explorer that asks what new interfaces AI enables.
We also tackle the “slop generator” problem and how to curb it. Different models have different tendencies, so route tasks where they fit: broad ideation to one, surgical changes to another. Constrain edits, ask for reasoning before code, and hand the model real context so it can propose focused steps. The same philosophy informs testing with computer-use models: if an agent can’t find your logout or complete checkout by looking at the UI, maybe your users struggle too. Rather than replacing developers, these tools elevate the craft—pushing commodity work downward while widening the canvas for design, problem framing, and tasteful implementation.
Want more? Check out ArtificialRuby.ai for upcoming events and videos, explore LatentLibrary.xyz, and find Scott’s essays and tutorials at WorksOnMyMachine.ai. If this conversation helps you rethink your workflow, follow, share with a teammate, and leave a review so more builders can join the experiment.
Honeybadger is an application health monitoring tool built by developers for developers.
Judoscale
Autoscaling that actually works. Take control of your cloud hosting.
Disclaimer: This post contains affiliate links. If you make a purchase, I may receive a commission at no extra cost to you.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
SPEAKER_01 (00:00):
Hello everyone and welcome to another episode of
Code and the Coding Coders toCoded.
I'm your host, Drew Bragg, andI'm joined today by Scott
Werner.
Scott, for anyone who'sunfamiliar with you, would you
please do a quick introduction?
SPEAKER_00 (00:12):
Sure.
Thanks, Drew.
So yeah, my name's Scott Werner.
I've been a Rubyist for sincearound 2008.
Worked in a bunch of differentplaces from small startups to
Groupon to Adobe, starting myown startup called Sayspring.
(00:33):
But now I'm at a new startupcalled Sublayer, and we're kind
of doing all different kinds ofstuff around AI and Ruby, which
is kind of how we got connected,right?
SPEAKER_01 (00:44):
You're working with AI.
What a shock.
I feel like everyone is in somecapacity, whether they're
working with it as their job orthey're working with it to help
them with productivity.
It is everywhere.
It's unavoidable.
So I'm glad we're going to talkabout it because it is something
that I've been shying away fromtalking about on the show, just
because there are times where itfeels a little bit like
(01:05):
JavaScript frameworks.
There's just something new withthem like every other day.
But before we get into theepisode, let me do the rest of
my normal spiel.
For anyone new to the show, I'mgoing to be asking Scott three
questions.
I'm going to ask him what he'sworking on, what kind of
blockers does he have?
And the last question is what'ssomething cool, new, or
interesting that he's recentlylearned, discovered, built?
(01:27):
It doesn't matter.
It's whatever he wants to share.
It's always my favoritequestion.
So let's just dive right in.
Scott, what are you working onspecifically?
SPEAKER_00 (01:38):
That's a tough question, as for anybody that
comes to subleyer.com might see.
We're doing kind of a big mix ofa bunch of different things.
You know, most of my day rightnow is spent.
So rewind a little bit.
So I started Sublayer back aboutthree, three and a half years
ago.
And the thing that kind ofkicked off what we're doing is
(02:01):
seeing GPT-4, like the demosthat they were putting out for
before GPT-4 came out, wherelike somebody would draw a UI
and it would write the code toactually like make it display.
And that point, seeing that,would made something click and
that like these things are justgoing to keep getting better.
These are like the worst thesemodels are ever going to be.
And it's going to fundamentallychange software engineering.
(02:23):
And I kind of brushed over itearlier, but I have been in
extreme programming, apractitioner of extreme
programming for almost over 10years now of pair programming,
test-driven development, agile.
And seeing that, it was like aspark went off in my head.
Everything that I knew or Ithought I knew is going to
(02:44):
change, and I want to beinvolved in that.
And winding road, probably threeepisodes worth of podcasts to
kind of talk about all thedifferent experiments we've
done.
But where I found myself nowdoing a mix of training for
agentic coding.
So using Claude Code, those kindof things, doing one-on-one and
small group training to kind ofteach different techniques and
(03:06):
get people up to speed.
(03:29):
Writing a newsletter works on mymachine, which kind of doing a
lot of the experiments inpublic, learning in public,
sharing a lot of open source,building products.
And then getting close tolaunching a video series kind of
very inspired by like Railscast,where it's like five to
15-minute really targetedtechniques for agentic coding or
(03:52):
using these tools.
On top of artificial Ruby andall these other things that
we're doing.
SPEAKER_01 (03:58):
Yeah, talk a little bit about Artificial Ruby for
anyone who's not familiar withthe meetup, because I think it
is pretty cool.
SPEAKER_00 (04:05):
So Artificial Ruby, it actually kind of came out
because as I was building outSublayer, a lot of the stuff we
were doing was in Ruby, andeverybody I talked to asked me
why Ruby.
Investors are like, why are youdoing like what's the
distinction here in Ruby?
Everybody, engineers, I wouldtalk to is like, isn't AI like a
(04:26):
Python and TypeScript kind ofthing?
And so I kind of got fed up as alittle bit of tough word, is
like a stronger word, but I itwas just like, you know what?
I'm sure there are more peopleout here doing this.
And so we actually put togethera happy hour where we're like, I
bet we can get like 20 of ourfriends to come out and we'll
buy them drinks and talk aboutRuby and AI.
(04:48):
So we did it and just shared ayou know, a Luma invite out and
got something like 50 people,and it was just like completely
like beyond our expectations.
And we were like, okay, there'slike something here.
In the middle of it, people wereasking us when the next one was.
And we were like, I don't know.
We just kind of thought it wasgonna be people we all knew, and
it's just like it expanded to somany new people.
(05:10):
So we did another one, and nowwe're about a year and a half
in.
It actually has a name because Ispoke at RubyConf last year, and
somebody came up to me and theywere like, Yeah, I heard there's
like this Ruby and AI event inNew York, but I can't find it.
And so we're like, oh, we needto actually like brand this and
not the Ruby and AI happy hourand demo night.
It's actually a name, all thisstuff.
(05:30):
It has a name, it's just kind ofa hard, it's sort of like my
podcast name.
SPEAKER_01 (05:33):
It's sort of hard to remember.
SPEAKER_00 (05:35):
But you actually rank really high, so that's the
difference.
Now we're artificial Ruby, wehave a site, we have a venue
that's great with beta works, avideo crew, we record talks.
It's actually kind of likebecome this thing where some
people have said it's almostlike a monthly mini conference,
and so it's kind of taken on alife of its own and just kind of
(05:56):
like guiding it as it drivesitself.
SPEAKER_01 (05:59):
Yeah, that's super cool.
I've been trying to carve outtime that I can come up because
it's not that far away from me,but it's just far enough away
that I can't just be like, todayI'm gonna go, why not?
It requires a little bit ofplanning.
SPEAKER_00 (06:14):
Yeah.
Ernesto has actually come up awhole bunch.
I got to know him.
Him and Ambu Labs are sponsors,actually.
So it's been cool to get to knowhim and then some of his team
and a couple people from histeam have given talks and
everything.
SPEAKER_01 (06:28):
If nothing else, I'll have you at Philly RB in
January, right?
We're still doing that.
Yes, yeah, definitely.
Yeah, that'll be cool.
We're getting that back going inperson, and it's been cool.
We've it's smaller for now, buteveryone we've done has had a
handful more people.
So it's nice to do the in-personmeetups monthly again.
Virtual is fun because you cantalk to anybody anywhere, but
(06:51):
there's definitely something toin-person that is a little more
special.
SPEAKER_00 (06:55):
Yeah, I agree.
I mean, it's one of the thingstoo that we've heard from
people, especially you know,from New York and meeting people
with artificial Ruby.
There are a bunch of people thathave started their career during
the pandemic and actually nevergone to an in-person
professional meetup at all, letalone a Ruby meetup.
And so, yeah, getting a chanceto like see people face to face.
(07:18):
Another tangent we can go on isgo imagine starting your
professional software developer,software engineering career
fully remote in the pandemic.
You've never seen any of yourcoworkers in person, and you put
your first PR up and it getstorn apart.
unknown (07:34):
Right?
SPEAKER_00 (07:34):
How hard it must be to be somebody like entering the
industry in like a remote orhybrid first place where you
kind of don't really get achance to make relationships
with people beyond like a nameor an avatar and Slack or
Discord.
SPEAKER_01 (07:49):
I was already in the industry, but I changed jobs
during the pandemic.
Went from one job to another,mostly because the job that I
went to was now remote and we'relike, oh, we're gonna stay
remote.
So now we can open up hiring.
So I was able to get a jobthere.
But it definitely was weird togo from I know what it's like to
(08:12):
work at a place, but like in anoffice, hybrid style, right?
Going to, I've never seen thesepeople.
And you know what the thing thatscrewed me up the most when I
finally did meet them, I had noone's height correct.
Everyone's height was way, I'mlike, you're shorter than I
thought, you're taller than Ithought.
Everyone's height was wrong.
(08:33):
That's what screwed me up themost, which is funny to think
about.
But yeah, I think the meetupsare great.
It's sort of like having theconferences back, right?
I've been saying, like, it'sgreat to have the conferences
back, it's great to have theconferences back, but it's also
great to have meetups,especially when you've got
cities like New York, Philly, SFRuby took off like crazy.
Just getting those peopletogether too helps because
(08:55):
sometimes those people don'ttravel to conferences.
So you wouldn't see themotherwise.
Sometimes those people are likenew to the industry or they've
been in it for a while andhaven't interacted with the
community.
Now there's a place to interactwith the community, and some
ideas that would never have seenthe light of day end up coming
out just because you're standingthere talking with the drink in
(09:17):
your hand and oh, tell me moreabout that.
And suddenly you're on this,it's 45 minutes to an hour
later, and you're like, this isawesome.
That doesn't happen over theinternet as easily.
It's more forced.
We tried, we tried for a littlewhile, right?
Do you remember the Zoop happyhours?
It was better than nothing, butgiven going to a meetup and
(09:38):
interacting with people inperson, and this is coming from
someone who has a limited socialbattery.
Like I get back from everyconference, like, I'm good, I
don't need to see any humanbeings for a few weeks now.
When it's once a month, it'sawesome.
Not looking to go back to anoffice, but I do like my
meetups.
Yeah, yeah.
So you've got your work, whichis AI focused, and you've got
(10:03):
your meetup, which is AIfocused.
So you are kind of one of the AIguys in Ruby.
What do you think is missingfrom the Ruby ecosystem to make
it where someone who's maybedoing AI work goes, well, I
gotta use Python becausePython's what you use for AI
(10:26):
work.
Someone who's doing AI work canlegitimately see Ruby as an
option to do AI work.
SPEAKER_00 (10:34):
That's something that I think about a lot and we
talk about a lot of the meetups.
But I think the time period,like the point in time that
we're in right now with AI, it'sso early, and nobody really
knows what we're actually goingto do with this.
If it stops at just likechatbots, just like an AI,
(10:55):
basically, it stops it atChatGPT.
I think it'll be a failure.
I think there's so many moreincredible things that we can do
with this technology.
What I think is missing, whatwe're trying to promote a lot
with artificial Ruby, is kind ofthe experimentation.
And so I think you look at whatdrove kind of the adoption of
(11:15):
Ruby in the early 2000s waspeople could make websites.
People were making web 2.0 appswith Java or PHP and stuff like
that before.
But it was Rails and somebodywith DHH, with the Ruby mindset
of like, I'm gonna approachthis, I'm gonna make this a joy
to work with.
(11:36):
That we saw the doors open andlike the flood of flood of
people come in.
I think where we're at right nowis like nobody really knows what
exactly we're gonna do with thisthing.
And so we don't have that lightbulb framework that's going to
be the thing, right?
I look back at that time period,and it looked like the Rubyists
(11:58):
were the ones having fun.
You know, I was listening toyour episode with Aji talking
about like the keynotes, and Iwasn't just like blowing smoke,
I've actually been doing a lotof trying to research like what
made it work in those early daysand why isn't it working now?
And you know, one of the thingsthat came up was one of DHH's
keynotes.
I think it was the keynote wherehe announced Rails, RubyConf.
(12:23):
And he was like, you need topick a fight, you need to create
this enemy.
And so it became enterprisedevelopers.
And like, oh, you're anenterprise developer, and like
the Rubyists were the ones withlike tattoos and different
colored hair and t-shirts andshorts, and they were the ones
that were like enjoying thework, and everybody else was
enterprise developers fillingout TPS reports.
(12:44):
And I've kind of gone way offtrack here.
There is no track.
There's no track, okay.
But so to bring it back, I thinkwe were in this kind of weird
period of probably something ofwhat it looked like before Rails
came out, where a lot of peoplejust found Ruby or been using
Ruby for a little while, werelike completely unlocked from
(13:07):
the things that they were doingbefore in their other languages,
in their other back then it wasthere weren't even like open
source IDEs really.
It was like the languages beingopen source was uncommon.
Now it's like everything'severybody expects everything to
be open source.
So I think my personal answerfor this, I don't think it's a
library, I don't think it's atool.
I think it's that feeling ofhaving fun.
(13:29):
And oh, that group of people,that group of programmers is
doing really cool stuff and theylook like they're having fun
with this.
I want to do that.
Because like Claude Code orwhatever your agent of choice is
can just spin up a copy of alibrary, you point it to
Langchain RB, and like do thisin Langchain.py.
And that might be a bad example.
(13:50):
But some library in Python, somelibrary in JavaScript, put it in
Claude or Gemini or whatever andbe like, make a copy of this in
Ruby, like it mostly will getyou there.
But I think it's getting back tothat culture of experimentation
and fun and just doing cool shitbecause you can.
SPEAKER_01 (14:06):
Yeah, I think that's a good way of looking at it.
Because we recently did a or aredoing, I guess, at the time of
this recording, a little bit oflike an AI challenge at work,
not hey, write all of your codeor vibe code anything, but more
of like go out of your way touse AI in your daily job as a
(14:30):
way of seeing like how are wegoing to use this here?
What is our AI usage look like?
Because almost across the board,with very few exceptions, like
none of us want AI doing ourjobs for us or writing our code
for us either.
Like, we enjoy our jobs, wedon't want AI to take it from
us.
(14:50):
But like to ignore that AIexists and just never use it
just seems like we're puttingourselves at a disadvantage to
other developers.
So the challenge is just to likeuse it a little bit more and
then submit the ways that you'veused it.
And I think it was such a goodway to unblock, at least
especially me, from AI, becauseit was such a well, if AI is
(15:14):
going to take my job, I don'twant to use it because fuck it.
Right?
Like, I don't like you.
You're taking this thing that Ilegitimately enjoy doing.
I like writing code, I likeprogramming, I like solving
problems.
Why would I want to just dump itinto this AI chatbot and let it
do all the stuff that I enjoy?
But because it was like arequirement to play with AI and
(15:35):
figure out ways to use it, I'vetried all different kinds of
things now and I haven't evenscratched the surface.
And I'm actually kind of havingfun with it.
I'm not using it to write code.
I'm using it to like explorethings or come up with plans.
I'm using it as like a rubberduck that talks back to me
instead of just sitting there onmy desk.
It's not writing any code, andI'm having so much more fun with
(15:57):
it than before when it was like,I guess I'll put it into GPT and
see what it comes up with, orI'll tap into Claude and be
like, I have this weird bug.
Those ways didn't seem fun, butjust playing with it seems like
so much fun.
SPEAKER_00 (16:13):
Do you have any like on top of mind, any kind of like
the biggest light bulb went offaha moment as you've been
working with it through thatproject?
SPEAKER_01 (16:21):
I don't think there's any one thing.
I've definitely learned a lotmore of like clawed code was
kind of cool when I firstinstalled it and like attached
my first MCP server to it.
We used linear to track bugs,and like the first time I was
able to like go back and forthwith it about a linear ticket
without like copying and pastingeverything in, that's kind of
(16:42):
cool.
I think there's something tothis connecting MCPs together.
The coolest thing that reallywas like an LLM on its own isn't
awesome.
It's a collection of thingsthat's pretty awesome.
Was I tried Tidewave by JoseValim, got that installed, and
started tinkering with it.
And just having it run in thebrowser and be able to like
(17:06):
click on something or talk aboutthe page as a whole, it was
leaps and bounds better thanClaude on its own.
I tried an experiment where Iwas like, hey, we have this
slow-loading page, Claude.
What do you think might be somelow-hanging fruits or areas to
explore?
Kept it pretty broad.
It came back with like aplan-ish places to look,
(17:28):
whatever.
I asked Tidewave, but becauseTidewave was in the browser, it
was reloading the page, it waslooking at my logs, it was
tapping into the database, itcame, it had all of these
contexts put together.
The plan it came up with, firstshot, was actually decent.
It would require some iterationto get it to be a really viable
(17:49):
plan, but it was just so muchbetter because it had so much
more context.
So, for at least for me, the ahamoment is like an LLM on its own
can only do so much.
But when you start giving it allof this context, it reduces my
cognitive load.
I don't need to keep all of thecontext in my head.
I can focus on solving problemsbecause the AI can keep all of
(18:10):
the context in its memory.
SPEAKER_00 (18:12):
Yeah, that's super cool.
I have TideWave downloaded it,but I haven't had a project to
really dig into it with, buthave to give that a shot.
SPEAKER_01 (18:18):
Yeah, it was cool.
Another one of the devs used itto fix like a CSS bug and had
decent success.
I've had mixed results with CSSbugs.
It's got like a selector tool,so you can select like a single
element and talk about thiselement isn't positioned
correctly or isn't the rightwhatever.
And it can do some pretty coolthings.
(18:40):
But I feel like every time AIjust is like, here's the 20
lines of CSS that'll fix it.
It gives me the light bulb togo, oh, and then like one or two
lines of CSS actually fixed it.
My God, this thing is a slopgenerator.
It fixed the bug.
It technically did what I askedit to do, but it did it in such
an overcomplicated like, I wouldnever want to be the one to
(19:03):
maintain this code way.
That's what I need to figure outis how to get it to stop going
way over what it needs to do.
Like, keep it simple.
I don't know how to convince theAI that like simple wins.
It just wants to write a shitton of code.
SPEAKER_00 (19:56):
Yeah.
I kind of have two answers tothat.
One is depends on which modelyou're working with.
Sonnet, for example, is one ofthe things that personality is
the wrong word, but like youkind of get a sense of like, I
imagine you've had thatexperience with Sonnet or Claude
Sonnet, because it's known tokind of like you give it a vague
instruction, it's gonna do fiveother things that seem
(20:19):
reasonable and just kind of likemake those up.
Codex, for example, from OpenAIwill be a lot more focused on
like just doing the thing thatyou're asking for.
And so you kind of start to getan intuition or start to like
kind of drive different problemsto different things.
So the library that I used forthe presentation at Rocky
Mountain Ruby, Monkey's Paw, itactually does amazing with
(20:43):
Sonnet because like I can justwrite a few lines.
So Monkey's Paw is thisprompt-based web framework.
And you kind of write yourslides, your pages in markdown
files in your wishes folder, andthen you load the page, it sends
that prompt to an LLM and loadswhatever it thinks should be on
that slide or page.
And it works way better withSonnet because I can just write
(21:07):
a couple lines and get this likewild the line with something the
style of something like amid-90s GeoCities site.
And I've got like animatedgradient backgrounds, I've got
these SVG diagrams, just likefrom two lines of like, I just
need a triangle or something.
That's very cool.
SPEAKER_01 (21:26):
Yeah, if anybody hasn't seen Scott's talk from
Rocky Mountain, I will link thevideo in the show notes.
You need to watch it.
It was great, and it is cool.
That talk was such a goodexample of like playing with
AIs, which was something that Iwasn't doing.
(21:46):
ChatGPT is just an LLM.
Like you ask it for stuff and itspits back out whatever
information it has.
But like seeing you do stuffwith Monkey Paul is like it felt
a little bit like some of Y'swork, which I'm not trying to
like shirk your ego here, butlike just the playing, the doing
(22:06):
something, not for oh, hey, I'mgonna make a new startup, or I'm
gonna make a real monetizableapplication, or I'm gonna make
something that like is in highdemand.
I'm not even gonna solvenecessarily a problem I have.
I'm just gonna build for thesake of building.
I'm gonna be creative, I'm gonnabe artsy.
And like, like that was wisework.
(22:27):
He was very creative, very artsywith Ruby.
And sometimes he createdsomething really cool that a lot
of people wanted to use.
And I'm sure a lot of stuff wasjust like this one-off, like, I
just want to see if I can do it.
And maybe that's not what youwanted to do with Monkey's Paw,
but either way, like that's kindof what I was like, this is so
(22:48):
cool.
Like, you just built thisframework because why not?
Because I wanted to see, Iwanted to iterate on something
for fun, wanted to do somethingcool with this tool and see what
I could do with it.
Which actually brings up anotherpoint.
If you the book I had it up andI forgot its name.
The latent library, oh, right,which is another one of those,
(23:13):
like, this is kind ofinteresting.
I'll let you do the talkingabout it, but I have so many
questions.
SPEAKER_00 (23:20):
Sure.
Completely embarrassed about thewhy comparison, but why was one
of the biggest personalitieswhen I was talking about like,
oh, the Ruby community wasthey're the ones having fun.
I want to go be around them.
Why was kind of the ringleaderfor that for me?
And he had a very big impact onmy entry into Ruby and into the
(23:44):
industry.
So definitely influenced the waythat I think about software.
Yeah.
And so a lot of these ideas arecoming out of just like, what
can I do now?
Not like, should I do it?
It's just like, what can I do?
And the other piece, and I kindof offhanded or kind of brushed
over this earlier, that like Ithink there's so much more than
(24:06):
just the chat interface andexploring that.
And so that's where library camefrom, which is a book social
network that none of the booksor categories actually exist.
They only exist by you browsing,and the LLM hallucinates the
categories and books for you.
And then once you check out thebook from the library, it
(24:28):
generates a cover, it becomesreal, and we're working toward
alternative interfaces foractually generating the content
of those books.
So you can have this kind ofcollaborative hallucination
authorship.
Almost like what does thismedium of an LLM or what does
this medium afford us that wehave this infinite creativity or
(24:48):
infinite any permutation ofcharacters up to the output
limit of an LLM is possible.
And so part of what LatentLibrary is trying to do is to
give you an interface for kindof navigating that infinity and
scoped toward books.
But the idea is like there areall these other interfaces that
(25:11):
we can do to find differentpermutations of these characters
that mean different things.
SPEAKER_01 (25:17):
It's like, what's that old give enough monkeys
typewriters, and eventuallysomeone one of them will write
Hamlet?
It's interesting to see likeyou're kind of like drilling
down this type of book, thattype of book, drill it down
until finally the AIhallucinates what you're looking
for.
I would have never thought to dothat.
(25:39):
And that's the cool, likeinteresting part for me when you
like shared it with me.
I was like, where does this ideacome from?
The creativity to me is not theoh, have AI generate a book,
right?
That's like that's actually notwhat I want AI to do.
I want AI to do my damn laundryso that I can go write the book.
(26:01):
That's where my headset's at.
But like you've kind of flippedit on the head.
It's like, yeah, I don't want AIto do this for me, but I have
this idea and I just want to seewhat happens.
Like, where in your brain doesthat come from?
SPEAKER_00 (26:12):
I don't know exactly.
I'm kind of just pullinginspiration from everywhere that
I think we're in this period oftime where so many new things
are possible that it's almostlike the internet came out, and
we're like, what can we do nowthat our machines are networked?
And so part of the inspirationfor even just starting the
(26:33):
newsletter was like, I kepthaving these ideas, but I'm also
working on a startup.
And so, like, I'm just like, youknow, just piling up these ideas
that I'm not able to work on.
And now through having thisoutlet, any random idea that I
have, I can just like throw it aclaw, clawed code, and just
like, yeah, build me out aprototype, build me out an
example, and let me see how thatis.
(26:54):
And actually, Monkey's Paw kindof I backed into it.
So my talked at RubyConf lastyear.
I had PowerPoint Copilot livegenerate my slides during my
talk.
And I kind of was thinking, I'mgonna take this further, and
then thought through all theproblems.
Like I was sitting there typingin prompts or copy and pasting
(27:14):
prompts while I'm up on behindthe podium.
And it was like, this sucks.
And so a lot of it was justlike, okay, like how do I
package that into a product?
I think it's just letting yourmind wander of just like, huh,
that thing's possible now.
What's something fun I can doabout this?
Or like, or this sucks, but howcan I make this not suck?
(27:35):
Latent library, it's I don'tremember what the spark was, but
it is really just one of thethings that I've found,
especially with Claude Code andyou know these other agenc
builders, is in the past, whenyou had an idea for something,
it was like, do I spend want tospend my weekends on this?
Do I want to spend the nextcouple weeks to like get this to
(27:56):
a state where I can share itwith some friends?
And now it's like I can send aprompt or send a couple prompts
to this agent and like do otherthings.
I can give Claude a prompt andgo do my laundry and come back
and check out what it lookslike, which is a big unlock that
it took a while of what you weretalking about that you guys were
doing at work of oh, I can dothis now.
(28:18):
What's something that I can haveAI do for me that I couldn't
before?
And just like pulling on thatthread and pulling on that
thread.
SPEAKER_01 (28:27):
Yeah, it's a little bit like the rails scaffolds,
right?
I wouldn't use what thescaffolds generate for me as the
final product, but it does makegetting like the boilerplate
crap out of the way really fastand like getting to the actual
fun, hard-to-solve problemsooner.
And that's where I actually wantto be.
(28:49):
I don't want to sit here andwrite the same boilerplate or
this is a controller, it theseare the actions it has, this is
a model, this it needs a schemamigration and all this.
Yeah, yeah, yeah, yeah, yeah.
I know how to do that.
I want to do the hard problembusiness y stuff, business
solution stuff.
That's where I want to be.
So AI sort of feels like that'skind of maybe where it fits in,
(29:11):
where it's like, these are thethings that we know how to do.
They don't take brain power.
Let the LLM do it.
Now let's go work on the realproblems.
Or at least I hope, becauseagain, I don't want AI to take
my job.
I like my job.
I enjoy what I do.
SPEAKER_00 (29:26):
Yeah, and I think Rails actually is a great
example of what becomes possiblewhen you take a lot of that
boilerplate stuff away.
Like, wouldn't have had Grouponor Twitter or Airbnb or Uber if
people weren't able to likequickly experiment with things.
SPEAKER_01 (29:43):
For sure.
I guess an interesting questionnow will be what do your
blockers look like?
Since you have this, like Ihesitate to call it a magic tool
to unblock you, but like itdefinitely is very good at
unblocking you.
If you have any kind of problem,like you can just dump it into
various LLMs chatbots, and itwill give you all kinds of ideas
(30:06):
on how to unblock you and stuff.
So, like, what does a blockerlook like now for you in your
startup or for artificial Ruby,whatever you want to talk about?
Do you think AI has made animpact in your blockers, or has
it just changed the type ofblockers you have?
SPEAKER_00 (30:23):
I think it's both.
I got to a point and gotcomfortable enough with
something like Cloud Code, whereI at one point had like four
running and Four differentprojects at the same time.
And so my blocker became justlike, I can't even listen to
music because I'm contextswitching so much that the music
is distracting.
(30:44):
It's definitely there are pointswhere, like, okay, it's not good
enough.
It's not good enough.
Our techniques aren't goodenough.
Our scaffolding isn't goodenough to do the thing I'm
trying to get it to do.
And so part of it's mental andlike personal of like, there's a
tendency, like, oh, I know howto do it.
I'm gonna put my time intocoding and doing that.
(31:07):
When in reality, like threemonths from now, an AI is gonna
be able to do it in a prompt.
So, like spending time doingthat is not really worthwhile.
And I think to I do an exercisevery similar to like what you
were talking about at work,where try to like get AI to do
the things for you.
And when you're going to dosomething, can AI do this?
And so I was procrastinating onsomething for a while.
(31:31):
Of I had Claude on Rails, whichis Obi Fernandez's, or I think
Obi Fernandez is involved inthis.
It's kind of like a multi-agentsystem for Claude to build Rails
apps.
And it got the thing working,but it was kind of convoluted
user flows.
And I was like, I don't want tolike dig through the routes and
(31:51):
like look through all thesethings and like clean this up.
And so it was just like puttingit off.
And a friend was like, why don'tyou just get Claude to do that?
I was like, oh my God, you'reright.
That morning, I was just like,hey, Claude, can you map out all
the routes and the primary userflows, the primary like paths
that users would do toaccomplish what they're trying
to do?
A couple minutes later, I hadthis whole like thing mapped
out, and it was like, okay,these look kind of convoluted.
(32:13):
Do you have any recommendationsfor simplifying it?
Okay, got a whole bunch ofrecommendations for simplifying
it.
Okay, do that.
And that was the thing that I'dbeen putting off for weeks,
actually sitting down and doingit.
A lot of the blockers are justretraining myself to be like,
oh, I should just take fiveminutes.
What's a new technique that Icould try to get AI to do this
rather than spending half a dayon something?
(32:35):
And then the other thing that'sI can get up to four clouds
running at the same time, but Ikind of run out of, I have the
ideas, but then you run out ofideas.
And so part of the motivationfor artificial Ruby, coming on
podcasts and the newsletter andspeaking conferences is really
to try to build a community ofpeople that we can toss ideas
(32:58):
around back and forth about.
I think that's the other blockerof just let's get together in a
chat room or a meetup and justlike riff on ideas because I
feel like we haven't evenscratched the surface of what
these things can do.
And really just trying to putstuff out there that people are
like, that's interesting.
I want to talk about this or Iwant to build stuff like this.
(33:19):
I want to take these ideas andtake them in a different
direction.
More people to chat with andjust play with ideas.
Can always use more.
SPEAKER_01 (33:28):
It's interesting too, because like that's I guess
the one thing that AI reallycan't do is it can't have an
idea.
It needs a prompt.
It needs to be responding tosomething nonstop.
When we as humans like get boredor we don't have anything to do,
like our brains just make upstuff.
You start humming a song thatdoesn't exist, or like whatever.
(33:51):
Maybe there is a future where AIgets to that point, but it's
kind of not there, right?
It needs human interaction inorder to do any kind of
generation.
ChatGPT just sits there and doesnothing until you ask it for
something.
I've been thinking about likehow weird it is.
You ask an LLM something, itjust puts a bunch of stuff
(34:14):
together and then spits it backat you in like this cohesive
summary.
It doesn't know anything.
It's connecting dots andsummarizing for you.
I guess the other thing is like,what's my role in AI code
generation or whatever?
It's like, is it just sittingthere making sure that the slop
(34:35):
generator doesn't generate toomuch slop?
Like, just enough slop to solvethe problem.
Like, don't create technicaldebt while you're solving the
problem.
Or is there something else thatI need to learn?
Like you keep on talking aboutlike retraining yourself.
Like, I'm trying to relearn howto do my job with AI involved in
(34:57):
a way that still brings me thejoy that I found programming and
writing code and solvingproblems, but like gives me the
productivity boost of I can askit to write a method for me, and
it can do that.
I don't really want it to, buthow can I get the productivity
boost of AI while still lovingmy job?
SPEAKER_00 (35:15):
It kind of looks like very similar to what we
were talking about before aroundRails with scaffolding and
everything kind of reduced a lotof the boilerplate.
I don't know, I don't think alot of people's arguments around
what they were doing in Javabefore Rails for web apps, that
they were like, the thing I loveabout building web apps in Java
is all the boilerplate.
(35:36):
It came out as the type system,how can you do anything without
types?
And like, almost like at thatpoint in time, it was like, oh,
the static types were solved andJava's solved programming, and
Rails and the Ruby world cameout and were like, yeah, but
what if we do the opposite ofall that and like don't even
have the code, you know, youcan't even look at the code
because it's all metaprogrammed.
(35:56):
And a lot of people had abacklash against that.
When what you got was fewerpeople were able to do so much
more, and the industry actuallyexpanded, like you ended up not
needing 100 person plus teamsand a bunch of stuff to build
something like Twitter, Groupon,or Airbnb, which in '98 you
(36:19):
probably did need hundreds ofpeople.
And I think we're seeing thatagain.
And so I think a lot of the crudapp stuff that we're moving
beyond just like the scaffold,we're moving to a lot of the
commodity work that we kind ofdo.
Then we've got the focus oncraft, and that is still going
(36:42):
to stay.
It's just like a lot of thestuff that we didn't really
categorize as boilerplate beforeis becoming boilerplate, and
we're just moving to anotherlayer.
I heard an analogy, I'm probablygoing to butcher this, but it's
almost like as photography cameout, and a lot of the people
that were focused on painting,painting portraits and
(37:04):
realistic, I don't know what theright term is, but realist
painting, that by photographycoming in and like taking the
economic, the commodity piece ofthat away, photography became
this whole other artisticmedium.
But it also kind of made itpossible for the people that
were focused on the craft ofpainting to like move into other
(37:25):
avenues and find otherinteresting things to do.
I don't know if we're actuallygoing to get there, but I think
the story is something likePicasso was an amazing painter,
like painting reality, but alsotook it in another direction and
was able to kind of learn allthe rules to break all the
rules.
My guess and the way I'm kind ofapproaching it is like we're
(37:46):
gonna see that same thing oflike a lot of the commodity
stuff that we spend our timedoing is gonna get taken away,
but like the craft of thebuilding, basically like the
creating software piece is stillgoing to exist.
The thing is, like, you know, Ireally like typing boilerplate
(38:59):
or typing curly brackets andparentheses.
That's probably going away.
I'm sorry about that.
But the creative problemsolving, I don't think is going
away.
SPEAKER_01 (39:09):
Yeah, good way to look at it.
So the last question, which Ifeel like is a bit of a loaded
one given some of the ideas thatyou've had, monkeys paws late in
library, things like that.
But like, what is somethingcool, new, or interesting?
Doesn't have to be AR-related,doesn't have to be coding
related, but it absolutely canbe.
Like, what's something cool,new, or interesting you want to
(39:30):
share that we haven't alreadytalked about?
SPEAKER_00 (39:33):
Well, I guess another new thing that I put out
that we haven't touched on, it'salso been bouncing around in my
head is Google recently put outthis model.
I think Anthropic has a computeruse model as well, but Google
put out a new computer use modelwhere you give it a screenshot
of a UI, it gives you back toolcalls of what to do, like click
(39:54):
at this X, Y coordinate or fillin this text field, whatever.
Another thing that I've recentlyreleased is called it Touring
Test, but it's a Cucumberextension that basically can
turn your cucumber steps intocomputer use steps.
So you describe what you wantthe agent to do, it'll do it,
(40:16):
and then you write yourassertions.
So if given I'm a logged inuser, your instruction is like
when the agent logs out, thenthey should be on the homepage
or whatever.
The agent logs out is the onlything that's there.
And so the agent sees your UI,it kind of determines where to
(40:37):
go and how to log out, and iteither does it or it doesn't.
And the thing I'm trying to getat there is like it almost
enables a new type of testingwhere your users are trying to
accomplish certain things inyour app.
And an LLM is a reasonable, it'snot gonna have all of the infer,
(40:57):
you know, all of theintelligence or all of the
abilities that your users might,but it's gonna take a, you know,
have a reasonable attempt atthese things.
And if it can't do it, it'sprobably a good sign that
there's some kind of usabilityissues on your site, which I
think is actually kind of whatwe're trying to get at with
testing in general, right?
We don't really care thatCapybara or or Selenium can
(41:20):
select this element with thisCSS class.
We're trying to test that theusers can add one of our
products to their cart and theycan check out and they can sign
up.
And there's the actual piece ofit.
There's the mechanical piece ofit, but then there's also the
can they actually figure out howto do it piece that wasn't
(41:41):
really there before.
SPEAKER_01 (41:42):
Interesting.
See, that's another use of AI.
I don't think I would havethought like I've thought of
obviously using AI to write my Rspec tests or to generate tests
for something it wrote or Iwrote or add tests so I could do
some TDD or what have you, butlike using it in a completely
different way.
(42:04):
It is such a weird, interesting.
Like, what are we gonna do withthis thing?
There's so many things it cando, but what the hell are we
gonna do with it?
I don't know.
You're definitely a lot furtherdown that path than I am.
I'm trying to catch up, I think,a little bit on the what are we
gonna do with this thing.
I'm happy that work did this AIchallenge because I do think it
(42:27):
took me from like AI to like,all right, I'm slower this month
because I'm playing with so muchstuff, but like it is definitely
going to improve my productivityin the long run.
It's going to stop me fromhaving to do some of the
boilerplatey stuff that was justa time suck.
Like, I could do it.
It wasn't hard.
(42:47):
I got knocked it out prettyquickly, but like definitely
faster with AI and will unlockme to solve bigger and better
problems.
SPEAKER_00 (42:57):
Yeah, I think that's that's a good point, too, is
I've had a lot of conversations,especially with like old
coworkers, of how to actually dothese things or any any tips for
this thing that I'm trying todo.
And a lot of people are like,you have deadlines, right?
I have to deliver this byFriday.
I'm not gonna noodle around withAI until Wednesday on the off
chance I'm late for thisdeliverable.
(43:19):
And I think that's anotherreally good thing that you have
there with that space toexperiment.
And I wonder if a lot morepeople than realize it have that
every company, every likeengineering manager, engineering
leader that I talk to is like, Iwant my team to be using AI
more, but I don't know how to doit.
(43:39):
And I wonder if it's just likeletting them know that if you
try to use AI to solve thisproblem and it takes a little
bit longer, that's okay.
SPEAKER_01 (43:49):
In a way, we're all kind of back to being juniors in
that space, right?
Like when we were juniorengineers, we couldn't produce
X, Y, or Z as fast, but weeventually became seniors.
And now we can do X, Y, and Zstupid fast, and we can work on
bigger, better problems.
But we need the time and spaceto like learn how to use the AI
(44:10):
and use the AI right for us.
There's 10 developers at Podia,and I think we all use AI
differently.
We all pretty much agree wedon't want to just vibe code our
PRs.
No one's doing that.
But we are all using AI in a waythat helps us, and we know that
in the long term we're going tobe more productive for it.
(44:30):
Having that space to explore is,yeah, you're right, is super
good.
And I think that might be ifthere's more holdouts, maybe
giving yourself the space ortalking to your manager about
getting the space to like mydeadline is pushed back, not so
I can slack off, but because Ican explore doing things
differently.
(44:51):
Because keep up or you're gonnabe left behind.
Cool.
That was a good one.
When is the next artificial Rubyfor anyone listening that's
like, hey, I'm in the NYC area,Philly area, anywhere that's
close enough to NYC to make thattrack.
When is the next one?
And do you have a schedule ofevents or is it kind of free
(45:14):
form?
How do they work?
SPEAKER_00 (45:16):
So the next one's December 3rd, and it's gonna be
the last one for the year.
Also, I've never really beeninvolved in any kind of event
planning whatsoever.
And so, like, I'm kind ofiterating and learning this as I
go.
We don't have a schedule ofevents yet.
I think we'll probably try toannounce the next one the night
(45:36):
of the event, but we're alsokind of like iterating.
So the best would be to go tothe site and sign up for the
mailing list.
We put out the information, andif you're not in the New York
area, we'll also send out emailsafterwards when the talks are up
on YouTube.
So nice.
SPEAKER_01 (45:52):
And that's artificialruby.com?
ArtificialRuby.ai, and I thinkthat makes sense.
Yeah, shocking.
Dot AI for the AI Ruby meetup.
Yep, that makes sense.
We also have the document.
So either one, either one.
Okay, there you go.
Yeah.
Covering the bases.
Domain addict, but aren't weall?
Isn't that a thing?
Isn't that like a prerequisiteof being a developer?
Is just owning domains andpotentially never doing anything
(46:15):
with them.
Although you are doing stuffwith them, so you're a step
above.
Any place on the internet that'sbest to find you specifically?
SPEAKER_00 (46:25):
Me specifically, subler.com will get you to kind
of everything we're doing.
Most of the writing that I do isworks on mymachine.ai.
I try to do it weekly.
Some days, sometimes if theproject's a little bit bigger,
it takes a couple weeks, butusually there.
And then from there, usuallyhanging out in the sublayer
(46:46):
Discord.
So if you chat, have some ideas,come there.
Cool.
SPEAKER_01 (46:50):
Very cool.
Well, is there anything wemissed before we do our wrap-up?
Like anything that you werehoping we'd talk about that we
didn't, or that you want to atleast touch on before we bounce?
SPEAKER_00 (47:02):
By the time this episode's published, latent
library will be should be live.
So that'll be at latentlibrary.xyz.
We can add that to the shownotes too.
And I think by then, you know, Imentioned earlier that I'm gonna
be putting out kind ofRailscasty style five to ten to
fifteen minute videos.
I think that should also belaunched by the time this
(47:25):
episode is published.
I'm gonna add that to the workson my machine substack.
So keep the essays and the opensource stuff and then add like
kind of a new$15 a month, twovideos a month of here's a
really focused technique onsolving a problem or doing
something with the agenticcoding, agenc programming
(47:48):
agents, like clock code, code I,yeah, Gemini.
SPEAKER_01 (47:51):
Do we call these things?
Yeah, naming is hard.
Everyone who's followed me doinganything ever should know I
struggle with naming.
Cool.
Well, thanks, Scott.
I really appreciate you comingon and talking about all that
cool AI stuff and lookingforward to having you in Philly.
And hopefully I can make it upto an artificial here soon,
because it definitely soundslike a cool place to talk to
(48:14):
people and find out more aboutthis very broad and weird space
of AI that has no clear path.
And that's both exciting and alittle daunting.
I do want to come up and checkit out because I got the Ruby
part down.
It's the AI part I'm stillfiguring out.
SPEAKER_00 (48:34):
Yeah, we'd love to have you.
It's definitely a great time.
And yeah, I'm excited to comedown to Philly and meet the
Philly crew too.
Yeah.
Yeah.
Thanks so much for having me on.
SPEAKER_01 (48:43):
This was a lot of fun.
Absolutely, man.
Check out Scott at all theplaces.
Check out the show notes forlinks, and I will see you all in
the next episode.
Bye.
Hello everyone and welcome to another episode of
Code and the Coding Coders toCoded.
I'm your host, Drew Bragg, andI'm joined today by Scott
Werner.
Scott, for anyone who'sunfamiliar with you, would you
please do a quick introduction?
SPEAKER_00 (00:12):
Sure.
Thanks, Drew.
So yeah, my name's Scott Werner.
I've been a Rubyist for sincearound 2008.
Worked in a bunch of differentplaces from small startups to
Groupon to Adobe, starting myown startup called Sayspring.
(00:33):
But now I'm at a new startupcalled Sublayer, and we're kind
of doing all different kinds ofstuff around AI and Ruby, which
is kind of how we got connected,right?
SPEAKER_01 (00:44):
You're working with AI.
What a shock.
I feel like everyone is in somecapacity, whether they're
working with it as their job orthey're working with it to help
them with productivity.
It is everywhere.
It's unavoidable.
So I'm glad we're going to talkabout it because it is something
that I've been shying away fromtalking about on the show, just
because there are times where itfeels a little bit like
(01:05):
JavaScript frameworks.
There's just something new withthem like every other day.
But before we get into theepisode, let me do the rest of
my normal spiel.
For anyone new to the show, I'mgoing to be asking Scott three
questions.
I'm going to ask him what he'sworking on, what kind of
blockers does he have?
And the last question is what'ssomething cool, new, or
interesting that he's recentlylearned, discovered, built?
(01:27):
It doesn't matter.
It's whatever he wants to share.
It's always my favoritequestion.
So let's just dive right in.
Scott, what are you working onspecifically?
SPEAKER_00 (01:38):
That's a tough question, as for anybody that
comes to subleyer.com might see.
We're doing kind of a big mix ofa bunch of different things.
You know, most of my day rightnow is spent.
So rewind a little bit.
So I started Sublayer back aboutthree, three and a half years
ago.
And the thing that kind ofkicked off what we're doing is
(02:01):
seeing GPT-4, like the demosthat they were putting out for
before GPT-4 came out, wherelike somebody would draw a UI
and it would write the code toactually like make it display.
And that point, seeing that,would made something click and
that like these things are justgoing to keep getting better.
These are like the worst thesemodels are ever going to be.
And it's going to fundamentallychange software engineering.
(02:23):
And I kind of brushed over itearlier, but I have been in
extreme programming, apractitioner of extreme
programming for almost over 10years now of pair programming,
test-driven development, agile.
And seeing that, it was like aspark went off in my head.
Everything that I knew or Ithought I knew is going to
(02:44):
change, and I want to beinvolved in that.
And winding road, probably threeepisodes worth of podcasts to
kind of talk about all thedifferent experiments we've
done.
But where I found myself nowdoing a mix of training for
agentic coding.
So using Claude Code, those kindof things, doing one-on-one and
small group training to kind ofteach different techniques and
(03:06):
get people up to speed.
(03:29):
Writing a newsletter works on mymachine, which kind of doing a
lot of the experiments inpublic, learning in public,
sharing a lot of open source,building products.
And then getting close tolaunching a video series kind of
very inspired by like Railscast,where it's like five to
15-minute really targetedtechniques for agentic coding or
(03:52):
using these tools.
On top of artificial Ruby andall these other things that
we're doing.
SPEAKER_01 (03:58):
Yeah, talk a little bit about Artificial Ruby for
anyone who's not familiar withthe meetup, because I think it
is pretty cool.
SPEAKER_00 (04:05):
So Artificial Ruby, it actually kind of came out
because as I was building outSublayer, a lot of the stuff we
were doing was in Ruby, andeverybody I talked to asked me
why Ruby.
Investors are like, why are youdoing like what's the
distinction here in Ruby?
Everybody, engineers, I wouldtalk to is like, isn't AI like a
(04:26):
Python and TypeScript kind ofthing?
And so I kind of got fed up as alittle bit of tough word, is
like a stronger word, but I itwas just like, you know what?
I'm sure there are more peopleout here doing this.
And so we actually put togethera happy hour where we're like, I
bet we can get like 20 of ourfriends to come out and we'll
buy them drinks and talk aboutRuby and AI.
(04:48):
So we did it and just shared ayou know, a Luma invite out and
got something like 50 people,and it was just like completely
like beyond our expectations.
And we were like, okay, there'slike something here.
In the middle of it, people wereasking us when the next one was.
And we were like, I don't know.
We just kind of thought it wasgonna be people we all knew, and
it's just like it expanded to somany new people.
(05:10):
So we did another one, and nowwe're about a year and a half
in.
It actually has a name because Ispoke at RubyConf last year, and
somebody came up to me and theywere like, Yeah, I heard there's
like this Ruby and AI event inNew York, but I can't find it.
And so we're like, oh, we needto actually like brand this and
not the Ruby and AI happy hourand demo night.
It's actually a name, all thisstuff.
(05:30):
It has a name, it's just kind ofa hard, it's sort of like my
podcast name.
SPEAKER_01 (05:33):
It's sort of hard to remember.
SPEAKER_00 (05:35):
But you actually rank really high, so that's the
difference.
Now we're artificial Ruby, wehave a site, we have a venue
that's great with beta works, avideo crew, we record talks.
It's actually kind of likebecome this thing where some
people have said it's almostlike a monthly mini conference,
and so it's kind of taken on alife of its own and just kind of
(05:56):
like guiding it as it drivesitself.
SPEAKER_01 (05:59):
Yeah, that's super cool.
I've been trying to carve outtime that I can come up because
it's not that far away from me,but it's just far enough away
that I can't just be like, todayI'm gonna go, why not?
It requires a little bit ofplanning.
SPEAKER_00 (06:14):
Yeah.
Ernesto has actually come up awhole bunch.
I got to know him.
Him and Ambu Labs are sponsors,actually.
So it's been cool to get to knowhim and then some of his team
and a couple people from histeam have given talks and
everything.
SPEAKER_01 (06:28):
If nothing else, I'll have you at Philly RB in
January, right?
We're still doing that.
Yes, yeah, definitely.
Yeah, that'll be cool.
We're getting that back going inperson, and it's been cool.
We've it's smaller for now, buteveryone we've done has had a
handful more people.
So it's nice to do the in-personmeetups monthly again.
Virtual is fun because you cantalk to anybody anywhere, but
(06:51):
there's definitely something toin-person that is a little more
special.
SPEAKER_00 (06:55):
Yeah, I agree.
I mean, it's one of the thingstoo that we've heard from
people, especially you know,from New York and meeting people
with artificial Ruby.
There are a bunch of people thathave started their career during
the pandemic and actually nevergone to an in-person
professional meetup at all, letalone a Ruby meetup.
And so, yeah, getting a chanceto like see people face to face.
(07:18):
Another tangent we can go on isgo imagine starting your
professional software developer,software engineering career
fully remote in the pandemic.
You've never seen any of yourcoworkers in person, and you put
your first PR up and it getstorn apart.
unknown (07:34):
Right?
SPEAKER_00 (07:34):
How hard it must be to be somebody like entering the
industry in like a remote orhybrid first place where you
kind of don't really get achance to make relationships
with people beyond like a nameor an avatar and Slack or
Discord.
SPEAKER_01 (07:49):
I was already in the industry, but I changed jobs
during the pandemic.
Went from one job to another,mostly because the job that I
went to was now remote and we'relike, oh, we're gonna stay
remote.
So now we can open up hiring.
So I was able to get a jobthere.
But it definitely was weird togo from I know what it's like to
(08:12):
work at a place, but like in anoffice, hybrid style, right?
Going to, I've never seen thesepeople.
And you know what the thing thatscrewed me up the most when I
finally did meet them, I had noone's height correct.
Everyone's height was way, I'mlike, you're shorter than I
thought, you're taller than Ithought.
Everyone's height was wrong.
(08:33):
That's what screwed me up themost, which is funny to think
about.
But yeah, I think the meetupsare great.
It's sort of like having theconferences back, right?
I've been saying, like, it'sgreat to have the conferences
back, it's great to have theconferences back, but it's also
great to have meetups,especially when you've got
cities like New York, Philly, SFRuby took off like crazy.
Just getting those peopletogether too helps because
(08:55):
sometimes those people don'ttravel to conferences.
So you wouldn't see themotherwise.
Sometimes those people are likenew to the industry or they've
been in it for a while andhaven't interacted with the
community.
Now there's a place to interactwith the community, and some
ideas that would never have seenthe light of day end up coming
out just because you're standingthere talking with the drink in
(09:17):
your hand and oh, tell me moreabout that.
And suddenly you're on this,it's 45 minutes to an hour
later, and you're like, this isawesome.
That doesn't happen over theinternet as easily.
It's more forced.
We tried, we tried for a littlewhile, right?
Do you remember the Zoop happyhours?
It was better than nothing, butgiven going to a meetup and
(09:38):
interacting with people inperson, and this is coming from
someone who has a limited socialbattery.
Like I get back from everyconference, like, I'm good, I
don't need to see any humanbeings for a few weeks now.
When it's once a month, it'sawesome.
Not looking to go back to anoffice, but I do like my
meetups.
Yeah, yeah.
So you've got your work, whichis AI focused, and you've got
(10:03):
your meetup, which is AIfocused.
So you are kind of one of the AIguys in Ruby.
What do you think is missingfrom the Ruby ecosystem to make
it where someone who's maybedoing AI work goes, well, I
gotta use Python becausePython's what you use for AI
(10:26):
work.
Someone who's doing AI work canlegitimately see Ruby as an
option to do AI work.
SPEAKER_00 (10:34):
That's something that I think about a lot and we
talk about a lot of the meetups.
But I think the time period,like the point in time that
we're in right now with AI, it'sso early, and nobody really
knows what we're actually goingto do with this.
If it stops at just likechatbots, just like an AI,
(10:55):
basically, it stops it atChatGPT.
I think it'll be a failure.
I think there's so many moreincredible things that we can do
with this technology.
What I think is missing, whatwe're trying to promote a lot
with artificial Ruby, is kind ofthe experimentation.
And so I think you look at whatdrove kind of the adoption of
(11:15):
Ruby in the early 2000s waspeople could make websites.
People were making web 2.0 appswith Java or PHP and stuff like
that before.
But it was Rails and somebodywith DHH, with the Ruby mindset
of like, I'm gonna approachthis, I'm gonna make this a joy
to work with.
(11:36):
That we saw the doors open andlike the flood of flood of
people come in.
I think where we're at right nowis like nobody really knows what
exactly we're gonna do with thisthing.
And so we don't have that lightbulb framework that's going to
be the thing, right?
I look back at that time period,and it looked like the Rubyists
(11:58):
were the ones having fun.
You know, I was listening toyour episode with Aji talking
about like the keynotes, and Iwasn't just like blowing smoke,
I've actually been doing a lotof trying to research like what
made it work in those early daysand why isn't it working now?
And you know, one of the thingsthat came up was one of DHH's
keynotes.
I think it was the keynote wherehe announced Rails, RubyConf.
(12:23):
And he was like, you need topick a fight, you need to create
this enemy.
And so it became enterprisedevelopers.
And like, oh, you're anenterprise developer, and like
the Rubyists were the ones withlike tattoos and different
colored hair and t-shirts andshorts, and they were the ones
that were like enjoying thework, and everybody else was
enterprise developers fillingout TPS reports.
(12:44):
And I've kind of gone way offtrack here.
There is no track.
There's no track, okay.
But so to bring it back, I thinkwe were in this kind of weird
period of probably something ofwhat it looked like before Rails
came out, where a lot of peoplejust found Ruby or been using
Ruby for a little while, werelike completely unlocked from
(13:07):
the things that they were doingbefore in their other languages,
in their other back then it wasthere weren't even like open
source IDEs really.
It was like the languages beingopen source was uncommon.
Now it's like everything'severybody expects everything to
be open source.
So I think my personal answerfor this, I don't think it's a
library, I don't think it's atool.
I think it's that feeling ofhaving fun.
(13:29):
And oh, that group of people,that group of programmers is
doing really cool stuff and theylook like they're having fun
with this.
I want to do that.
Because like Claude Code orwhatever your agent of choice is
can just spin up a copy of alibrary, you point it to
Langchain RB, and like do thisin Langchain.py.
And that might be a bad example.
(13:50):
But some library in Python, somelibrary in JavaScript, put it in
Claude or Gemini or whatever andbe like, make a copy of this in
Ruby, like it mostly will getyou there.
But I think it's getting back tothat culture of experimentation
and fun and just doing cool shitbecause you can.
SPEAKER_01 (14:06):
Yeah, I think that's a good way of looking at it.
Because we recently did a or aredoing, I guess, at the time of
this recording, a little bit oflike an AI challenge at work,
not hey, write all of your codeor vibe code anything, but more
of like go out of your way touse AI in your daily job as a
(14:30):
way of seeing like how are wegoing to use this here?
What is our AI usage look like?
Because almost across the board,with very few exceptions, like
none of us want AI doing ourjobs for us or writing our code
for us either.
Like, we enjoy our jobs, wedon't want AI to take it from
us.
(14:50):
But like to ignore that AIexists and just never use it
just seems like we're puttingourselves at a disadvantage to
other developers.
So the challenge is just to likeuse it a little bit more and
then submit the ways that you'veused it.
And I think it was such a goodway to unblock, at least
especially me, from AI, becauseit was such a well, if AI is
(15:14):
going to take my job, I don'twant to use it because fuck it.
Right?
Like, I don't like you.
You're taking this thing that Ilegitimately enjoy doing.
I like writing code, I likeprogramming, I like solving
problems.
Why would I want to just dump itinto this AI chatbot and let it
do all the stuff that I enjoy?
But because it was like arequirement to play with AI and
(15:35):
figure out ways to use it, I'vetried all different kinds of
things now and I haven't evenscratched the surface.
And I'm actually kind of havingfun with it.
I'm not using it to write code.
I'm using it to like explorethings or come up with plans.
I'm using it as like a rubberduck that talks back to me
instead of just sitting there onmy desk.
It's not writing any code, andI'm having so much more fun with
(15:57):
it than before when it was like,I guess I'll put it into GPT and
see what it comes up with, orI'll tap into Claude and be
like, I have this weird bug.
Those ways didn't seem fun, butjust playing with it seems like
so much fun.
SPEAKER_00 (16:13):
Do you have any like on top of mind, any kind of like
the biggest light bulb went offaha moment as you've been
working with it through thatproject?
SPEAKER_01 (16:21):
I don't think there's any one thing.
I've definitely learned a lotmore of like clawed code was
kind of cool when I firstinstalled it and like attached
my first MCP server to it.
We used linear to track bugs,and like the first time I was
able to like go back and forthwith it about a linear ticket
without like copying and pastingeverything in, that's kind of
(16:42):
cool.
I think there's something tothis connecting MCPs together.
The coolest thing that reallywas like an LLM on its own isn't
awesome.
It's a collection of thingsthat's pretty awesome.
Was I tried Tidewave by JoseValim, got that installed, and
started tinkering with it.
And just having it run in thebrowser and be able to like
(17:06):
click on something or talk aboutthe page as a whole, it was
leaps and bounds better thanClaude on its own.
I tried an experiment where Iwas like, hey, we have this
slow-loading page, Claude.
What do you think might be somelow-hanging fruits or areas to
explore?
Kept it pretty broad.
It came back with like aplan-ish places to look,
(17:28):
whatever.
I asked Tidewave, but becauseTidewave was in the browser, it
was reloading the page, it waslooking at my logs, it was
tapping into the database, itcame, it had all of these
contexts put together.
The plan it came up with, firstshot, was actually decent.
It would require some iterationto get it to be a really viable
(17:49):
plan, but it was just so muchbetter because it had so much
more context.
So, for at least for me, the ahamoment is like an LLM on its own
can only do so much.
But when you start giving it allof this context, it reduces my
cognitive load.
I don't need to keep all of thecontext in my head.
I can focus on solving problemsbecause the AI can keep all of
(18:10):
the context in its memory.
SPEAKER_00 (18:12):
Yeah, that's super cool.
I have TideWave downloaded it,but I haven't had a project to
really dig into it with, buthave to give that a shot.
SPEAKER_01 (18:18):
Yeah, it was cool.
Another one of the devs used itto fix like a CSS bug and had
decent success.
I've had mixed results with CSSbugs.
It's got like a selector tool,so you can select like a single
element and talk about thiselement isn't positioned
correctly or isn't the rightwhatever.
And it can do some pretty coolthings.
(18:40):
But I feel like every time AIjust is like, here's the 20
lines of CSS that'll fix it.
It gives me the light bulb togo, oh, and then like one or two
lines of CSS actually fixed it.
My God, this thing is a slopgenerator.
It fixed the bug.
It technically did what I askedit to do, but it did it in such
an overcomplicated like, I wouldnever want to be the one to
(19:03):
maintain this code way.
That's what I need to figure outis how to get it to stop going
way over what it needs to do.
Like, keep it simple.
I don't know how to convince theAI that like simple wins.
It just wants to write a shitton of code.
SPEAKER_00 (19:56):
Yeah.
I kind of have two answers tothat.
One is depends on which modelyou're working with.
Sonnet, for example, is one ofthe things that personality is
the wrong word, but like youkind of get a sense of like, I
imagine you've had thatexperience with Sonnet or Claude
Sonnet, because it's known tokind of like you give it a vague
instruction, it's gonna do fiveother things that seem
(20:19):
reasonable and just kind of likemake those up.
Codex, for example, from OpenAIwill be a lot more focused on
like just doing the thing thatyou're asking for.
And so you kind of start to getan intuition or start to like
kind of drive different problemsto different things.
So the library that I used forthe presentation at Rocky
Mountain Ruby, Monkey's Paw, itactually does amazing with
(20:43):
Sonnet because like I can justwrite a few lines.
So Monkey's Paw is thisprompt-based web framework.
And you kind of write yourslides, your pages in markdown
files in your wishes folder, andthen you load the page, it sends
that prompt to an LLM and loadswhatever it thinks should be on
that slide or page.
And it works way better withSonnet because I can just write
(21:07):
a couple lines and get this likewild the line with something the
style of something like amid-90s GeoCities site.
And I've got like animatedgradient backgrounds, I've got
these SVG diagrams, just likefrom two lines of like, I just
need a triangle or something.
That's very cool.
SPEAKER_01 (21:26):
Yeah, if anybody hasn't seen Scott's talk from
Rocky Mountain, I will link thevideo in the show notes.
You need to watch it.
It was great, and it is cool.
That talk was such a goodexample of like playing with
AIs, which was something that Iwasn't doing.
(21:46):
ChatGPT is just an LLM.
Like you ask it for stuff and itspits back out whatever
information it has.
But like seeing you do stuffwith Monkey Paul is like it felt
a little bit like some of Y'swork, which I'm not trying to
like shirk your ego here, butlike just the playing, the doing
(22:06):
something, not for oh, hey, I'mgonna make a new startup, or I'm
gonna make a real monetizableapplication, or I'm gonna make
something that like is in highdemand.
I'm not even gonna solvenecessarily a problem I have.
I'm just gonna build for thesake of building.
I'm gonna be creative, I'm gonnabe artsy.
And like, like that was wisework.
(22:27):
He was very creative, very artsywith Ruby.
And sometimes he createdsomething really cool that a lot
of people wanted to use.
And I'm sure a lot of stuff wasjust like this one-off, like, I
just want to see if I can do it.
And maybe that's not what youwanted to do with Monkey's Paw,
but either way, like that's kindof what I was like, this is so
(22:48):
cool.
Like, you just built thisframework because why not?
Because I wanted to see, Iwanted to iterate on something
for fun, wanted to do somethingcool with this tool and see what
I could do with it.
Which actually brings up anotherpoint.
If you the book I had it up andI forgot its name.
The latent library, oh, right,which is another one of those,
(23:13):
like, this is kind ofinteresting.
I'll let you do the talkingabout it, but I have so many
questions.
SPEAKER_00 (23:20):
Sure.
Completely embarrassed about thewhy comparison, but why was one
of the biggest personalitieswhen I was talking about like,
oh, the Ruby community wasthey're the ones having fun.
I want to go be around them.
Why was kind of the ringleaderfor that for me?
And he had a very big impact onmy entry into Ruby and into the
(23:44):
industry.
So definitely influenced the waythat I think about software.
Yeah.
And so a lot of these ideas arecoming out of just like, what
can I do now?
Not like, should I do it?
It's just like, what can I do?
And the other piece, and I kindof offhanded or kind of brushed
over this earlier, that like Ithink there's so much more than
(24:06):
just the chat interface andexploring that.
And so that's where library camefrom, which is a book social
network that none of the booksor categories actually exist.
They only exist by you browsing,and the LLM hallucinates the
categories and books for you.
And then once you check out thebook from the library, it
(24:28):
generates a cover, it becomesreal, and we're working toward
alternative interfaces foractually generating the content
of those books.
So you can have this kind ofcollaborative hallucination
authorship.
Almost like what does thismedium of an LLM or what does
this medium afford us that wehave this infinite creativity or
(24:48):
infinite any permutation ofcharacters up to the output
limit of an LLM is possible.
And so part of what LatentLibrary is trying to do is to
give you an interface for kindof navigating that infinity and
scoped toward books.
But the idea is like there areall these other interfaces that
(25:11):
we can do to find differentpermutations of these characters
that mean different things.
SPEAKER_01 (25:17):
It's like, what's that old give enough monkeys
typewriters, and eventuallysomeone one of them will write
Hamlet?
It's interesting to see likeyou're kind of like drilling
down this type of book, thattype of book, drill it down
until finally the AIhallucinates what you're looking
for.
I would have never thought to dothat.
(25:39):
And that's the cool, likeinteresting part for me when you
like shared it with me.
I was like, where does this ideacome from?
The creativity to me is not theoh, have AI generate a book,
right?
That's like that's actually notwhat I want AI to do.
I want AI to do my damn laundryso that I can go write the book.
(26:01):
That's where my headset's at.
But like you've kind of flippedit on the head.
It's like, yeah, I don't want AIto do this for me, but I have
this idea and I just want to seewhat happens.
Like, where in your brain doesthat come from?
SPEAKER_00 (26:12):
I don't know exactly.
I'm kind of just pullinginspiration from everywhere that
I think we're in this period oftime where so many new things
are possible that it's almostlike the internet came out, and
we're like, what can we do nowthat our machines are networked?
And so part of the inspirationfor even just starting the
(26:33):
newsletter was like, I kepthaving these ideas, but I'm also
working on a startup.
And so, like, I'm just like, youknow, just piling up these ideas
that I'm not able to work on.
And now through having thisoutlet, any random idea that I
have, I can just like throw it aclaw, clawed code, and just
like, yeah, build me out aprototype, build me out an
example, and let me see how thatis.
(26:54):
And actually, Monkey's Paw kindof I backed into it.
So my talked at RubyConf lastyear.
I had PowerPoint Copilot livegenerate my slides during my
talk.
And I kind of was thinking, I'mgonna take this further, and
then thought through all theproblems.
Like I was sitting there typingin prompts or copy and pasting
(27:14):
prompts while I'm up on behindthe podium.
And it was like, this sucks.
And so a lot of it was justlike, okay, like how do I
package that into a product?
I think it's just letting yourmind wander of just like, huh,
that thing's possible now.
What's something fun I can doabout this?
Or like, or this sucks, but howcan I make this not suck?
(27:35):
Latent library, it's I don'tremember what the spark was, but
it is really just one of thethings that I've found,
especially with Claude Code andyou know these other agenc
builders, is in the past, whenyou had an idea for something,
it was like, do I spend want tospend my weekends on this?
Do I want to spend the nextcouple weeks to like get this to
(27:56):
a state where I can share itwith some friends?
And now it's like I can send aprompt or send a couple prompts
to this agent and like do otherthings.
I can give Claude a prompt andgo do my laundry and come back
and check out what it lookslike, which is a big unlock that
it took a while of what you weretalking about that you guys were
doing at work of oh, I can dothis now.
(28:18):
What's something that I can haveAI do for me that I couldn't
before?
And just like pulling on thatthread and pulling on that
thread.
SPEAKER_01 (28:27):
Yeah, it's a little bit like the rails scaffolds,
right?
I wouldn't use what thescaffolds generate for me as the
final product, but it does makegetting like the boilerplate
crap out of the way really fastand like getting to the actual
fun, hard-to-solve problemsooner.
And that's where I actually wantto be.
(28:49):
I don't want to sit here andwrite the same boilerplate or
this is a controller, it theseare the actions it has, this is
a model, this it needs a schemamigration and all this.
Yeah, yeah, yeah, yeah, yeah.
I know how to do that.
I want to do the hard problembusiness y stuff, business
solution stuff.
That's where I want to be.
So AI sort of feels like that'skind of maybe where it fits in,
(29:11):
where it's like, these are thethings that we know how to do.
They don't take brain power.
Let the LLM do it.
Now let's go work on the realproblems.
Or at least I hope, becauseagain, I don't want AI to take
my job.
I like my job.
I enjoy what I do.
SPEAKER_00 (29:26):
Yeah, and I think Rails actually is a great
example of what becomes possiblewhen you take a lot of that
boilerplate stuff away.
Like, wouldn't have had Grouponor Twitter or Airbnb or Uber if
people weren't able to likequickly experiment with things.
SPEAKER_01 (29:43):
For sure.
I guess an interesting questionnow will be what do your
blockers look like?
Since you have this, like Ihesitate to call it a magic tool
to unblock you, but like itdefinitely is very good at
unblocking you.
If you have any kind of problem,like you can just dump it into
various LLMs chatbots, and itwill give you all kinds of ideas
(30:06):
on how to unblock you and stuff.
So, like, what does a blockerlook like now for you in your
startup or for artificial Ruby,whatever you want to talk about?
Do you think AI has made animpact in your blockers, or has
it just changed the type ofblockers you have?
SPEAKER_00 (30:23):
I think it's both.
I got to a point and gotcomfortable enough with
something like Cloud Code, whereI at one point had like four
running and Four differentprojects at the same time.
And so my blocker became justlike, I can't even listen to
music because I'm contextswitching so much that the music
is distracting.
(30:44):
It's definitely there are pointswhere, like, okay, it's not good
enough.
It's not good enough.
Our techniques aren't goodenough.
Our scaffolding isn't goodenough to do the thing I'm
trying to get it to do.
And so part of it's mental andlike personal of like, there's a
tendency, like, oh, I know howto do it.
I'm gonna put my time intocoding and doing that.
(31:07):
When in reality, like threemonths from now, an AI is gonna
be able to do it in a prompt.
So, like spending time doingthat is not really worthwhile.
And I think to I do an exercisevery similar to like what you
were talking about at work,where try to like get AI to do
the things for you.
And when you're going to dosomething, can AI do this?
And so I was procrastinating onsomething for a while.
(31:31):
Of I had Claude on Rails, whichis Obi Fernandez's, or I think
Obi Fernandez is involved inthis.
It's kind of like a multi-agentsystem for Claude to build Rails
apps.
And it got the thing working,but it was kind of convoluted
user flows.
And I was like, I don't want tolike dig through the routes and
(31:51):
like look through all thesethings and like clean this up.
And so it was just like puttingit off.
And a friend was like, why don'tyou just get Claude to do that?
I was like, oh my God, you'reright.
That morning, I was just like,hey, Claude, can you map out all
the routes and the primary userflows, the primary like paths
that users would do toaccomplish what they're trying
to do?
A couple minutes later, I hadthis whole like thing mapped
out, and it was like, okay,these look kind of convoluted.
(32:13):
Do you have any recommendationsfor simplifying it?
Okay, got a whole bunch ofrecommendations for simplifying
it.
Okay, do that.
And that was the thing that I'dbeen putting off for weeks,
actually sitting down and doingit.
A lot of the blockers are justretraining myself to be like,
oh, I should just take fiveminutes.
What's a new technique that Icould try to get AI to do this
rather than spending half a dayon something?
(32:35):
And then the other thing that'sI can get up to four clouds
running at the same time, but Ikind of run out of, I have the
ideas, but then you run out ofideas.
And so part of the motivationfor artificial Ruby, coming on
podcasts and the newsletter andspeaking conferences is really
to try to build a community ofpeople that we can toss ideas
(32:58):
around back and forth about.
I think that's the other blockerof just let's get together in a
chat room or a meetup and justlike riff on ideas because I
feel like we haven't evenscratched the surface of what
these things can do.
And really just trying to putstuff out there that people are
like, that's interesting.
I want to talk about this or Iwant to build stuff like this.
(33:19):
I want to take these ideas andtake them in a different
direction.
More people to chat with andjust play with ideas.
Can always use more.
SPEAKER_01 (33:28):
It's interesting too, because like that's I guess
the one thing that AI reallycan't do is it can't have an
idea.
It needs a prompt.
It needs to be responding tosomething nonstop.
When we as humans like get boredor we don't have anything to do,
like our brains just make upstuff.
You start humming a song thatdoesn't exist, or like whatever.
(33:51):
Maybe there is a future where AIgets to that point, but it's
kind of not there, right?
It needs human interaction inorder to do any kind of
generation.
ChatGPT just sits there and doesnothing until you ask it for
something.
I've been thinking about likehow weird it is.
You ask an LLM something, itjust puts a bunch of stuff
(34:14):
together and then spits it backat you in like this cohesive
summary.
It doesn't know anything.
It's connecting dots andsummarizing for you.
I guess the other thing is like,what's my role in AI code
generation or whatever?
It's like, is it just sittingthere making sure that the slop
(34:35):
generator doesn't generate toomuch slop?
Like, just enough slop to solvethe problem.
Like, don't create technicaldebt while you're solving the
problem.
Or is there something else thatI need to learn?
Like you keep on talking aboutlike retraining yourself.
Like, I'm trying to relearn howto do my job with AI involved in
(34:57):
a way that still brings me thejoy that I found programming and
writing code and solvingproblems, but like gives me the
productivity boost of I can askit to write a method for me, and
it can do that.
I don't really want it to, buthow can I get the productivity
boost of AI while still lovingmy job?
SPEAKER_00 (35:15):
It kind of looks like very similar to what we
were talking about before aroundRails with scaffolding and
everything kind of reduced a lotof the boilerplate.
I don't know, I don't think alot of people's arguments around
what they were doing in Javabefore Rails for web apps, that
they were like, the thing I loveabout building web apps in Java
is all the boilerplate.
(35:36):
It came out as the type system,how can you do anything without
types?
And like, almost like at thatpoint in time, it was like, oh,
the static types were solved andJava's solved programming, and
Rails and the Ruby world cameout and were like, yeah, but
what if we do the opposite ofall that and like don't even
have the code, you know, youcan't even look at the code
because it's all metaprogrammed.
(35:56):
And a lot of people had abacklash against that.
When what you got was fewerpeople were able to do so much
more, and the industry actuallyexpanded, like you ended up not
needing 100 person plus teamsand a bunch of stuff to build
something like Twitter, Groupon,or Airbnb, which in '98 you
(36:19):
probably did need hundreds ofpeople.
And I think we're seeing thatagain.
And so I think a lot of the crudapp stuff that we're moving
beyond just like the scaffold,we're moving to a lot of the
commodity work that we kind ofdo.
Then we've got the focus oncraft, and that is still going
(36:42):
to stay.
It's just like a lot of thestuff that we didn't really
categorize as boilerplate beforeis becoming boilerplate, and
we're just moving to anotherlayer.
I heard an analogy, I'm probablygoing to butcher this, but it's
almost like as photography cameout, and a lot of the people
that were focused on painting,painting portraits and
(37:04):
realistic, I don't know what theright term is, but realist
painting, that by photographycoming in and like taking the
economic, the commodity piece ofthat away, photography became
this whole other artisticmedium.
But it also kind of made itpossible for the people that
were focused on the craft ofpainting to like move into other
(37:25):
avenues and find otherinteresting things to do.
I don't know if we're actuallygoing to get there, but I think
the story is something likePicasso was an amazing painter,
like painting reality, but alsotook it in another direction and
was able to kind of learn allthe rules to break all the
rules.
My guess and the way I'm kind ofapproaching it is like we're
(37:46):
gonna see that same thing oflike a lot of the commodity
stuff that we spend our timedoing is gonna get taken away,
but like the craft of thebuilding, basically like the
creating software piece is stillgoing to exist.
The thing is, like, you know, Ireally like typing boilerplate
(38:59):
or typing curly brackets andparentheses.
That's probably going away.
I'm sorry about that.
But the creative problemsolving, I don't think is going
away.
SPEAKER_01 (39:09):
Yeah, good way to look at it.
So the last question, which Ifeel like is a bit of a loaded
one given some of the ideas thatyou've had, monkeys paws late in
library, things like that.
But like, what is somethingcool, new, or interesting?
Doesn't have to be AR-related,doesn't have to be coding
related, but it absolutely canbe.
Like, what's something cool,new, or interesting you want to
(39:30):
share that we haven't alreadytalked about?
SPEAKER_00 (39:33):
Well, I guess another new thing that I put out
that we haven't touched on, it'salso been bouncing around in my
head is Google recently put outthis model.
I think Anthropic has a computeruse model as well, but Google
put out a new computer use modelwhere you give it a screenshot
of a UI, it gives you back toolcalls of what to do, like click
(39:54):
at this X, Y coordinate or fillin this text field, whatever.
Another thing that I've recentlyreleased is called it Touring
Test, but it's a Cucumberextension that basically can
turn your cucumber steps intocomputer use steps.
So you describe what you wantthe agent to do, it'll do it,
(40:16):
and then you write yourassertions.
So if given I'm a logged inuser, your instruction is like
when the agent logs out, thenthey should be on the homepage
or whatever.
The agent logs out is the onlything that's there.
And so the agent sees your UI,it kind of determines where to
(40:37):
go and how to log out, and iteither does it or it doesn't.
And the thing I'm trying to getat there is like it almost
enables a new type of testingwhere your users are trying to
accomplish certain things inyour app.
And an LLM is a reasonable, it'snot gonna have all of the infer,
(40:57):
you know, all of theintelligence or all of the
abilities that your users might,but it's gonna take a, you know,
have a reasonable attempt atthese things.
And if it can't do it, it'sprobably a good sign that
there's some kind of usabilityissues on your site, which I
think is actually kind of whatwe're trying to get at with
testing in general, right?
We don't really care thatCapybara or or Selenium can
(41:20):
select this element with thisCSS class.
We're trying to test that theusers can add one of our
products to their cart and theycan check out and they can sign
up.
And there's the actual piece ofit.
There's the mechanical piece ofit, but then there's also the
can they actually figure out howto do it piece that wasn't
(41:41):
really there before.
SPEAKER_01 (41:42):
Interesting.
See, that's another use of AI.
I don't think I would havethought like I've thought of
obviously using AI to write my Rspec tests or to generate tests
for something it wrote or Iwrote or add tests so I could do
some TDD or what have you, butlike using it in a completely
different way.
(42:04):
It is such a weird, interesting.
Like, what are we gonna do withthis thing?
There's so many things it cando, but what the hell are we
gonna do with it?
I don't know.
You're definitely a lot furtherdown that path than I am.
I'm trying to catch up, I think,a little bit on the what are we
gonna do with this thing.
I'm happy that work did this AIchallenge because I do think it
(42:27):
took me from like AI to like,all right, I'm slower this month
because I'm playing with so muchstuff, but like it is definitely
going to improve my productivityin the long run.
It's going to stop me fromhaving to do some of the
boilerplatey stuff that was justa time suck.
Like, I could do it.
It wasn't hard.
(42:47):
I got knocked it out prettyquickly, but like definitely
faster with AI and will unlockme to solve bigger and better
problems.
SPEAKER_00 (42:57):
Yeah, I think that's that's a good point, too, is
I've had a lot of conversations,especially with like old
coworkers, of how to actually dothese things or any any tips for
this thing that I'm trying todo.
And a lot of people are like,you have deadlines, right?
I have to deliver this byFriday.
I'm not gonna noodle around withAI until Wednesday on the off
chance I'm late for thisdeliverable.
(43:19):
And I think that's anotherreally good thing that you have
there with that space toexperiment.
And I wonder if a lot morepeople than realize it have that
every company, every likeengineering manager, engineering
leader that I talk to is like, Iwant my team to be using AI
more, but I don't know how to doit.
(43:39):
And I wonder if it's just likeletting them know that if you
try to use AI to solve thisproblem and it takes a little
bit longer, that's okay.
SPEAKER_01 (43:49):
In a way, we're all kind of back to being juniors in
that space, right?
Like when we were juniorengineers, we couldn't produce
X, Y, or Z as fast, but weeventually became seniors.
And now we can do X, Y, and Zstupid fast, and we can work on
bigger, better problems.
But we need the time and spaceto like learn how to use the AI
(44:10):
and use the AI right for us.
There's 10 developers at Podia,and I think we all use AI
differently.
We all pretty much agree wedon't want to just vibe code our
PRs.
No one's doing that.
But we are all using AI in a waythat helps us, and we know that
in the long term we're going tobe more productive for it.
(44:30):
Having that space to explore is,yeah, you're right, is super
good.
And I think that might be ifthere's more holdouts, maybe
giving yourself the space ortalking to your manager about
getting the space to like mydeadline is pushed back, not so
I can slack off, but because Ican explore doing things
differently.
(44:51):
Because keep up or you're gonnabe left behind.
Cool.
That was a good one.
When is the next artificial Rubyfor anyone listening that's
like, hey, I'm in the NYC area,Philly area, anywhere that's
close enough to NYC to make thattrack.
When is the next one?
And do you have a schedule ofevents or is it kind of free
(45:14):
form?
How do they work?
SPEAKER_00 (45:16):
So the next one's December 3rd, and it's gonna be
the last one for the year.
Also, I've never really beeninvolved in any kind of event
planning whatsoever.
And so, like, I'm kind ofiterating and learning this as I
go.
We don't have a schedule ofevents yet.
I think we'll probably try toannounce the next one the night
(45:36):
of the event, but we're alsokind of like iterating.
So the best would be to go tothe site and sign up for the
mailing list.
We put out the information, andif you're not in the New York
area, we'll also send out emailsafterwards when the talks are up
on YouTube.
So nice.
SPEAKER_01 (45:52):
And that's artificialruby.com?
ArtificialRuby.ai, and I thinkthat makes sense.
Yeah, shocking.
Dot AI for the AI Ruby meetup.
Yep, that makes sense.
We also have the document.
So either one, either one.
Okay, there you go.
Yeah.
Covering the bases.
Domain addict, but aren't weall?
Isn't that a thing?
Isn't that like a prerequisiteof being a developer?
Is just owning domains andpotentially never doing anything
(46:15):
with them.
Although you are doing stuffwith them, so you're a step
above.
Any place on the internet that'sbest to find you specifically?
SPEAKER_00 (46:25):
Me specifically, subler.com will get you to kind
of everything we're doing.
Most of the writing that I do isworks on mymachine.ai.
I try to do it weekly.
Some days, sometimes if theproject's a little bit bigger,
it takes a couple weeks, butusually there.
And then from there, usuallyhanging out in the sublayer
(46:46):
Discord.
So if you chat, have some ideas,come there.
Cool.
SPEAKER_01 (46:50):
Very cool.
Well, is there anything wemissed before we do our wrap-up?
Like anything that you werehoping we'd talk about that we
didn't, or that you want to atleast touch on before we bounce?
SPEAKER_00 (47:02):
By the time this episode's published, latent
library will be should be live.
So that'll be at latentlibrary.xyz.
We can add that to the shownotes too.
And I think by then, you know, Imentioned earlier that I'm gonna
be putting out kind ofRailscasty style five to ten to
fifteen minute videos.
I think that should also belaunched by the time this
(47:25):
episode is published.
I'm gonna add that to the workson my machine substack.
So keep the essays and the opensource stuff and then add like
kind of a new$15 a month, twovideos a month of here's a
really focused technique onsolving a problem or doing
something with the agenticcoding, agenc programming
(47:48):
agents, like clock code, code I,yeah, Gemini.
SPEAKER_01 (47:51):
Do we call these things?
Yeah, naming is hard.
Everyone who's followed me doinganything ever should know I
struggle with naming.
Cool.
Well, thanks, Scott.
I really appreciate you comingon and talking about all that
cool AI stuff and lookingforward to having you in Philly.
And hopefully I can make it upto an artificial here soon,
because it definitely soundslike a cool place to talk to
(48:14):
people and find out more aboutthis very broad and weird space
of AI that has no clear path.
And that's both exciting and alittle daunting.
I do want to come up and checkit out because I got the Ruby
part down.
It's the AI part I'm stillfiguring out.
SPEAKER_00 (48:34):
Yeah, we'd love to have you.
It's definitely a great time.
And yeah, I'm excited to comedown to Philly and meet the
Philly crew too.
Yeah.
Yeah.
Thanks so much for having me on.
SPEAKER_01 (48:43):
This was a lot of fun.
Absolutely, man.
Check out Scott at all theplaces.
Check out the show notes forlinks, and I will see you all in
the next episode.
Bye.
Popular Podcasts
Las Culturistas with Matt Rogers and Bowen Yang
Ding dong! Join your culture consultants, Matt Rogers and Bowen Yang, on an unforgettable journey into the beating heart of CULTURE. Alongside sizzling special guests, they GET INTO the hottest pop-culture moments of the day and the formative cultural experiences that turned them into Culturistas. Produced by the Big Money Players Network and iHeartRadio.
Crime Junkie
Does hearing about a true crime case always leave you scouring the internet for the truth behind the story? Dive into your next mystery with Crime Junkie. Every Monday, join your host Ashley Flowers as she unravels all the details of infamous and underreported true crime cases with her best friend Brit Prawat. From cold cases to missing persons and heroes in our community who seek justice, Crime Junkie is your destination for theories and stories you won’t hear anywhere else. Whether you're a seasoned true crime enthusiast or new to the genre, you'll find yourself on the edge of your seat awaiting a new episode every Monday. If you can never get enough true crime... Congratulations, you’ve found your people. Follow to join a community of Crime Junkies! Crime Junkie is presented by Audiochuck Media Company.
The Brothers Ortiz
The Brothers Ortiz is the story of two brothers–both successful, but in very different ways. Gabe Ortiz becomes a third-highest ranking officer in all of Texas while his younger brother Larry climbs the ranks in Puro Tango Blast, a notorious Texas Prison gang. Gabe doesn’t know all the details of his brother’s nefarious dealings, and he’s made a point not to ask, to protect their relationship. But when Larry is murdered during a home invasion in a rented beach house, Gabe has no choice but to look into what happened that night. To solve Larry’s murder, Gabe, and the whole Ortiz family, must ask each other tough questions.