πŸ“Ί Develpreneur YouTube Episode

Video + transcript

How Vibe Coding with AI Is Changing the Way Developers Build Software

2025-10-28 β€’Youtube

Detailed Notes

In this episode of Building Better Foundations, Rob Broadhead and Michael Meloche explore vibe coding β€” the emerging trend of using AI to code in real time. Learn how developers are blending human creativity with AI precision to build faster, smarter, and more efficiently.

Discover the pros, cons, and real-world tips for making vibe coding part of your workflow β€” without falling into the trap of technical debt or confirmation bias.

πŸ‘‰ Topics Covered: β€’ What vibe coding really is β€’ How AI speeds up software development β€’ Using user stories and TDD with AI β€’ Avoiding technical debt β€’ Tips for smarter, cleaner coding

🎧 Listen to the full Building Better Developers podcast at https://develpreneur.com/vibe-coding-is-changing-the-way-developers-build-software/

*Follow-us on:*

* [email protected] * https://develpreneur.com/ * https://www.youtube.com/@develpreneur * https://facebook.com/Develpreneur * https://x.com/develpreneur * https://www.linkedin.com/company/develpreneur/

βΈ» #VibeCoding #AICoding #SoftwareDevelopment #BuildingBetterDevelopers #Develpreneur

Transcript Text
record because
this time I hit record. And welcome back
everybody. We are not doing an interview
this time around.
We are doing foundational stuff and I
have no clue what we want to do it on.
Um,
>> you didn't share that list you did last
time, but we had uh we've done the no
code, low code, uh, AI.
Uh, we've had the interview about trust
and reliability, marketing and remote
hiring.
>> Oh, I didn't uh Yeah, let's see. Um,
>> done. That was podcast ideas.
Uh
uh that's an interview.
Well, for foundational stuff, should we
do maybe one on um like project
planning? Maybe code, not so much code
reviews, but like code repositories,
code like keeping track of code.
Just trying to throw some ideas out.
>> Let me see. I'm trying to see.
Let me see if I can get back what I had
last done because we did talk about like
foundational stuff.
Let me pull up the blog for the We did
no code, low code.
>> Did we do vibe coding?
>> Uh, I think we were maybe going to do
that. We talked about it, but I don't
think we got too far in detail on it.
>> I don't think
we did.
Um,
>> well, let's do one on vod coding then
and go from there. Um,
>> yeah, because we talked about AI because
we just mentioned it, but we really
didn't do it. So, let's dive into that.
Um, and talk a little bit about
foundational stuff on that
and maybe how to do it right. Let me
see. Make sure I've got everything lined
up here.
And then there, get myself over here.
We're just going to go right into a
three, two, there. That's a two, a one.
Hello and welcome back. We are
continuing our season where we are
building better foundations. Actually
though we are building better developers
the developer podcast. I am one of the
hosts. I am Rob Broadhead. I'm a founder
of developer also a founder of RB
Consulting where we help you do
technology better. We have this thing
called a technology assessment. It's
sort of like a a checkup for your car or
a financial audit or uh actually when I
say audit, usually people don't like the
a word, but more of a financial
assessment that you might do with your
advisor. Same thing. Think of us as your
technology advisor. We sit down, help
first understand what you do, what your
business is, what are your visions, what
are your goals, and then we walk through
a process that allows us to create a
technology roadmap for you. Talk about
things that you can do better. And that
may be simplifying, integrating, even
innovating. We may even, you know, help
you build something new or uh just
migrating to new systems and upgrade
upgrades, things of that nature. Take
that big investment of technology and
make it work for you. So now it is
actually turnurning out revenue. We have
that roadmap for success. You can check
us out at rb-sns.com.
Uh we've got a couple different products
and projects and all kinds of little
things going out there. So check it out
and let us know what you think. Good
thing, bad thing. Um,
bad. Well, let's see. Boy, there's so
many good things and bad things. So,
good thing is, uh, modern flooring is
not very hard to put together. If you
get like the little snap in and stuff
like that, it can look really good,
really fast, uh, really professional.
Bad thing is, because I've been working
on this, one, I am not a DIY kind of
person. I will go figure that stuff out,
but I'm just I I don't do good jobs with
stuff. I'm the person that's like not
likely to, you know, get a nail hammered
straight in often enough or uh, you
know, getting an edge exactly, you know,
measured correct and all that kind of
stuff. And so dealing with things like
this where you're like off by a
millimeter and you notice it is a little
bit frustrating for me. So that has been
a bad thing. But it's going to get
better because for the next little bit
I'm going to be doing a podcast instead
of working on flooring. And I'm going to
allow you guys to be introduced to
Michael on the other side.
>> Hey everyone, my name is Mike Malash.
I'm one of the co-founders building
better developers, also known as
developer. I'm also the founder and
owner of Envision QA where we help
businesses take control of their
software by focusing on what matters and
that is quality, reliability, and
support that you can count on. Whether
you're building something new or trying
to fix what's broken, we combine custom
development with thorough testing to
make sure your systems actually work.
This way before you get in in front of
your customers and let them touch it. We
make sure that it works for them. Uh at
Envision QA, we don't just build
software. We make sure it's solid,
stable, and ready to support your
business growth. Check us out at
envisionqa.com.
Uh good thing, bad thing. Uh I'll start
with the bad thing. Um, of course,
weather is getting colder and I get a
call from my daughter this morning that
their HVAC went out. So, yet again,
another household thing has to be
replaced that is not a low ticket item.
Uh, good thing weather is getting
cooler, trees are starting to change
color, and we're starting to get into
fall. So, I'm starting to love it.
>> It is good. Uh, that is the nice thing
about the areas that we live in that uh
you get four seasons. So, you do
actually get to see like the leaves
change and stuff like that. Of course,
that also stirs up allergies, but that's
another discussion. Today, we're going
to dive back into we talked about low
code and no code as our foundational
stuff and we wanted we touched on AI and
vibe coding and I think we're going to
do a little bit more of that uh today
because we really didn't have the time
to uh you know give it the proper uh
attention essentially.
Now vibe coding uh let's start with the
idea of vibe coding is that you are
working through uh solving a problem and
you're using AI to generate some code
and give you and basically allowing AI
to sort of code as you go and there's
there are a lot of apps that are I guess
apps we'll put quotes around those
there's a lot of things that have been
built out there a lot of people I've
seen that are uh essentially putting
together MVPs minimally viable products
of some sort using vibe coding Now the
problem with vibe coding is that there
is also a huge amount of technology debt
that ends up involved there because AI
is not going to have that cohesive
thought that you need as far as as
building software. So from a
foundational point of view I want to
talk about so you know if if vibe coding
is in itself maybe not good um then how
do we find a way to make it good enough
so that it can help us be more
productive and that's where I want to
focus in in my discussion is
finding is like anything else and this
goes back to what I've said before um I
got from somebody else actually is that
when you're dealing with AI think about
it as dealing with a very junior maybe
mid-level, you know, in this case
developer. That means you have to be
very specific.
So if you're building an app, let's say
using vibe coding, it can be very
efficient. A good example is if you
build a page that is part of your
application and it g it has a look and a
feel and a a style and a structure and u
you know the backend infrastructure and
architecture is sort of there or
actually not even sort of is there and
now you're just building on top of it
then AI can be very good as long as you
say given this architecture given this
style given this look and feel I need
you to build me a page that does X and
it's not going to be 100% but it can get
you very quickly to a lot of the code
that you normally would be like you know
maybe copy and pasting or you're like
just typing the stuff that you've typed
a thousand times before
and if it's and the more you can refine
and define your request the better it's
going to be. So there's a lot of times
that I will, you know, if I want to
build a page very quickly, like a new
page for an application, I can, I guess
we'll call it vibe code, but basically
I'm going to use AI and say here's my
CSS. Here's a page that's almost exactly
what I want. Here's the data model.
These are the fields that I want to
display. This is how I want to display
them. You know, this is the general look
and feel. Then it will spit something
back out fairly quickly. you know, maybe
in minutes it will give me something
that it would have taken me maybe 20 or
30 minutes or longer to code out.
Particularly if it's something complex,
if it's something where you've got a lot
of of of lists or a lot of very
repetitive things where you can say,
"Oh, hey, I want this page." And it's
going to have 10 sections in it, it can
throw up that page very quickly. And
especially if you just need like that uh
that skeleton essentially that you're
going to then flesh out. And that
actually is the next thing that I would
say is like keep your request small and
maybe start with uh particularly in a
conversation start with an outline.
Start with something that you can start
you know a skeleton that you can start
putting the pieces together. So if you
think of a page I'm give you an example
before I pass this on to Michael and his
thoughts. If you have a page let's start
with like think of a web application
you're going to have uh some sort of
navigation. So, if you know what the
navigation roughly looks like, you know,
maybe you want to do a top menu, you
want to do a side menu, maybe you want
to do pop out menus, you want to do a
menu off the right, the left, the
bottom, however it is.
Once you can describe that,
you can say, "Hey, give me a menu that
follows these that, you know, follows
these points that actually hits these
requirements." And it may not always
give you all the requirements. It may
not hit all of them, but it's going to
give you something now that you can work
with. And so what you could do is you
could take that and maybe it doesn't and
you say, "Oh, wait a minute. These two
navigation items need to be a different
color." And then you can specifically
drill down into it and say, "Okay, let's
make these a different color." And this
is, I know, a very simple example, but
this is a way that you can actually vibe
code your way into being more
productive. You have to start with a
good foundation. Now, you can use the
vibe piece to start to put those pieces
together for the foundation, but if you
don't want to have a lot of technical
debt, you're going to have to go back
and clean it up as you go and make sure
that it is, you know, what you what you
are producing is clean and then also
give it back clean because usually if
you give it back, if you give it
something in a certain style that
follows certain standards, you're going
to see something that is similar to
those standards. And then be careful to
make sure that if it kicks back more
than you ask for that you don't pull in
that extra croft basically because a lot
of times it's going to end up being some
sort of technical debt. That's like I
know that was more of a soap box than an
intro. Uh but I will step off my soap
box and then toss that over to you
Michael your thoughts.
>> Yeah so typically I look at vibe coding.
So given my background with testing and
you know the focus of what I've been
doing for the last 10 years is more
test-driven development vibe coding to
me can be very successful especially if
you come at it from a testers
perspective or at least a user story
perspective. So like you said you got to
be very descriptive. You have to kind of
walk AI through what it is that you
want. The thing is if you actually walk
it through with clear user stories like
hey as a user I need to uh log into an
application. I need a login page and you
walk through kind of the requirements
that you need for the page and then AI
can generate. Now you can be very kind
of clean and straightforward like you
said like here add a button here and
things like that. But if you walk
through it through requirements focus
through the actual user story, you
actually can get better clarification
with AI as far as feedback. The other
thing I found is as you go through the
interaction with AI to build the
application from a user story
perspective,
you can actually start out by saying,
"Hey, write me a test case to do this."
And it'll build you a test case. And
then you can kind of then walk from the
test case into the application like okay
build me a test case to test a login
page. Okay now build me a login page for
this test case now and then you copy
paste this into your application. What I
find that's really useful with these
user stories especially with coding and
for vibe coding is you can eliminate a
lot of that boilerplate or at least the
structural understanding of code. So, if
you're not a coder and you don't really
know a language, you can just start out
with, hey, how do I build a website?
What do I need to build a website? And
it will say, okay, well, you need X, Y,
and Z. Okay, I don't understand that.
Ask it. Okay, what how do I set this up?
And
really,
AI can get you almost there.
There are times where it will get you
off the rails. You may need to go find
another AI engine or just jump out to
Google and search. Okay, AI tell me I
need to build a web server. All right,
it's giving me some steps, but I'm
having problems. AI is not quite getting
me there. So, take where you're
struggling and drop it out into Google.
Yes, Google still does some AI as well,
but you might get some better results
through Google to help you unplug some
of those issues. So, V coding can get
you mostly there, especially if you
don't know anything about coding. But
getting it to build you something that
you want or help you understand what you
want. It is better to go at it from a to
me go at better from a user story
perspective and walk it through what you
want like be very clear because user
stories really translate to requirements
and if you can explain it to AI clearly
and flush it out the confusion it has I
think you're going to get your a better
result of vibe coding with AI than if
you don't.
I agree that there I think that's that
is probably the key is getting the user
stories right. Um the problem with those
or I guess the challenge with that when
you do this is to make sure that you
also take those to the right uh the
right level. uh it's very easy uh and
this is a lot of what we do in
requirements gathering and I think
that'll probably be maybe the next
episode is we'll talk about that a
little bit
is we tend to we start with a big story
usually like I love to use you know the
ATM example of like I just want an ATM
where I have an automated teller well
cool but then what does that teller do
what are the features are going to be
available okay I want to be able to
deposit money and I want to be able to
withdraw money okay well is it for one
account? Is it for multiple accounts? Is
it how would you do that? Because before
an AP ATM, it's like, well, nobody had
an ATM card. So, how would that work?
What happens? And then there's the all
the questions around like, what happens
if I try to withdraw money and there's
no money in the machine or if I don't
have enough money in my account or what
if I try to deposit money and I don't
actually deposit the cash or I don't
deposit the check. How do I validate
that check? How do I validate the cash?
How do I count the thing? There's
there's a lot of these things that goes
on and on and on. And so when you're
building your stories is
it works again great to start from like
sort of the high level but then you need
to build it into a much more detailed
outline. it is and you have to
you have to sort of check yourself
because I have found too often that
people say like and I've had those
conversations with customers not AI but
just actually
human beings human eye um where it's you
know saying well this is what I want and
as you have a conversation you realize
that what they said they want is not
actually what they want um there's a lot
lot of people will hear things like uh
CRM or ERP or uh EMR or all these other
like tools and systems that are out
there and they'll say like so and so
said I need this I need an HR system and
you find out that it's a somebody that
somebody that has zero employees and
like uh I don't know if you really need
a HR system. What do you actually what
problem are you actually trying to
solve? And I think too often we start
with the story instead of the problem. I
think if you take it back a step and say
this is the problem. This is the
description of what I think the solution
looks like then a lot of times then
you'll be able to get some stuff back
that says well you'll maybe AI will ask
you question say well what about this or
what about that or it may say that
sounds like this kind of a solution.
Should I look up some more about that?
Should I educate you on that? And those
are the ways that I think that you can
step into uh AI and vibe coding and help
yourself get to a solution that is
actually a little more useful with a
little less uh overhead and some of the
extra stuff that you might have to clean
up after the fact.
I think we will I'll give you one more
because we're we don't want to go too
long because we're going to actually
spoiler alert we're going to spill into
some of this in the next episode.
Thoughts?
>> So I I kind of want to touch on the
negatives of this a little bit. So if
you're going into AI and you you're
asking those things or you you have a
problem but you haven't really defined
the problem, be careful with AI vibe
coding or even asking AI for too much
input into what you need really to go
talk to your customer about because you
could get what's called confirmation
bias. You could essentially go down a
track of AI is telling you what you
think you need and you're basically
feeding AI. Oh, okay. Tell me more about
and you're going down the wrong rabbit
hole and you're actually going away from
the problem or you're going away from
the correct requirement. So, make sure
that you do kind of check yourself as
you're going through this and even check
with the customer to make sure that
you've got the right requirement before
you go too far down building a system.
That's a good Yeah, it's uh when in
doubt, check yourself. Particularly when
you're in an area where you don't um you
don't have domain knowledge, where
you're not really sure what you're
doing. There's too often uh this goes
back to and you then we'll wrap this one
up. I think the thought is like there's
been a lot of times that we've had
customers, we've complained about this
where or potential customers, prospects
that have said, "Hey, I want to build um
their project definition. I want to
build eBay for uh honey or something
like that or I want to be the Amazon for
uh flower delivery or something like
that and that's it. That's like their
story. That's not going to cut it. Those
kinds of things like you need to really
spend some time researching and thinking
about what it is you're getting into.
And a lot of times I think you really
need to go look when you get those
answers. Do not take those answers
because AI is still based on the
internet basically. Don't take those
answers as gospel because there's a lot
of times you're going to get an answer
and when you really start looking at it
um you're going to find out that it's
wrong. Uh I don't know how many times
I've I've corrected AI and it's like oh
yeah you're right. I didn't think about
that or I wasn't using that or I was
thinking this or you know I was focusing
on this other thing sometimes just like
human beings.
But the direct is always going to be the
best way. things like send us an email
at infoddeveloper.com and let us know
your thoughts. What do you what are your
thoughts? I I would love to hear like
low code no code vibe coding stories. Uh
I know a lot of people are big on I see
a lot of people that are projects out
there that are potential customers like
we want to use this, we want to no code
this and low code that and vibe code
that other thing. And I'd like to see
some people that are like particularly
that have been successful that have
taken one of those systems completely to
production and served, you know, more
than five people at a time or something
like that. And I've seen some, I'm not
saying they don't exist. I've built a
couple of things that are in these
realms that have done that. But I would
love to hear some other stories because
it seems like they are few and far
between. Most of them seem like they are
things where people are like, "Okay, we
built this thing in low no code. We vibe
coded it. now we have to rewrite it
completely in something else because
they really didn't get they just got a
like a play solution and not something
that was production ready which is often
the case with uh code generation tools.
I digress again but uh check us out
developer.com you can leave us comments
there. We've got plenty and plenty of
content there. The YouTube channel
developer YouTube channel we've got all
of like multiple seasons of episodes not
to mention a lot of other stuff out
there.
Um, you can check us out on X. We are at
Developer. We have a Facebook page.
We're all of those good things. So,
wherever you wherever you subscribe to
podcast, leave us a review, leave us
some feedback. We would love to hear it
and uh help you out wherever we can with
new, you know, new seasons, new
episodes, new topics that are the ones
that you requested because we're here
for you. That being said, I have said
too much. Go out there and have yourself
a great day, a great week, and we will
talk to you next time.
Uh, bonus materials. You got to see the
genesis of this this episode. So, we're
going to keep this one a little bit
short because
you know what? You can't always have
bonus. It's not Christmas every day. You
don't get bonus after every episode. And
depending on how this next one goes, you
may get a lot of bonus stuff. Thank you
so much for your time. We do appreciate
you. Appreciate you guys hanging out
here. uh all the ways I said give us the
feedback you know it's all the feels
it's all the feeds all the feedbacks uh
let us know and we would be love love
love to hear from you if you want to be
a you know on the show let us know uh
we're open to discussions with a lot of
different people for our conversations
our interviews and uh we're more than
happy to bring people of all stripes
because that is what makes us all better
go out and have yourself a great Today,
Transcript Segments
27.119

record because

29.199

this time I hit record. And welcome back

31.679

everybody. We are not doing an interview

34.96

this time around.

37.12

We are doing foundational stuff and I

40.16

have no clue what we want to do it on.

44.559

Um,

48

>> you didn't share that list you did last

49.68

time, but we had uh we've done the no

52.079

code, low code, uh, AI.

55.52

Uh, we've had the interview about trust

57.36

and reliability, marketing and remote

60.559

hiring.

62.8

>> Oh, I didn't uh Yeah, let's see. Um,

70.799

>> done. That was podcast ideas.

75.52

Uh

82.56

uh that's an interview.

88.32

Well, for foundational stuff, should we

90.24

do maybe one on um like project

93.92

planning? Maybe code, not so much code

97.28

reviews, but like code repositories,

99.6

code like keeping track of code.

103.68

Just trying to throw some ideas out.

105.84

>> Let me see. I'm trying to see.

116.56

Let me see if I can get back what I had

118.56

last done because we did talk about like

122.079

foundational stuff.

131.039

Let me pull up the blog for the We did

133.84

no code, low code.

137.68

>> Did we do vibe coding?

141.2

>> Uh, I think we were maybe going to do

143.04

that. We talked about it, but I don't

144.48

think we got too far in detail on it.

147.12

>> I don't think

151.28

we did.

153.519

Um,

157.519

>> well, let's do one on vod coding then

159.68

and go from there. Um,

161.84

>> yeah, because we talked about AI because

163.68

we just mentioned it, but we really

165.2

didn't do it. So, let's dive into that.

168.16

Um, and talk a little bit about

170.4

foundational stuff on that

173.68

and maybe how to do it right. Let me

176.239

see. Make sure I've got everything lined

179.2

up here.

181.04

And then there, get myself over here.

184.159

We're just going to go right into a

185.519

three, two, there. That's a two, a one.

189.519

Hello and welcome back. We are

191.68

continuing our season where we are

193.44

building better foundations. Actually

195.04

though we are building better developers

196.56

the developer podcast. I am one of the

199.84

hosts. I am Rob Broadhead. I'm a founder

202

of developer also a founder of RB

204.08

Consulting where we help you do

206.4

technology better. We have this thing

208.56

called a technology assessment. It's

210.4

sort of like a a checkup for your car or

213.2

a financial audit or uh actually when I

215.92

say audit, usually people don't like the

217.36

a word, but more of a financial

218.879

assessment that you might do with your

220.72

advisor. Same thing. Think of us as your

223.44

technology advisor. We sit down, help

226.159

first understand what you do, what your

228.48

business is, what are your visions, what

230

are your goals, and then we walk through

232.64

a process that allows us to create a

234.799

technology roadmap for you. Talk about

237.2

things that you can do better. And that

238.64

may be simplifying, integrating, even

241.439

innovating. We may even, you know, help

243.04

you build something new or uh just

245.12

migrating to new systems and upgrade

247.28

upgrades, things of that nature. Take

249.2

that big investment of technology and

251.68

make it work for you. So now it is

253.68

actually turnurning out revenue. We have

256

that roadmap for success. You can check

258.079

us out at rb-sns.com.

260.32

Uh we've got a couple different products

262.24

and projects and all kinds of little

264.08

things going out there. So check it out

265.6

and let us know what you think. Good

268

thing, bad thing. Um,

272.56

bad. Well, let's see. Boy, there's so

274.56

many good things and bad things. So,

276.56

good thing is, uh, modern flooring is

280.4

not very hard to put together. If you

282.16

get like the little snap in and stuff

283.759

like that, it can look really good,

286.4

really fast, uh, really professional.

289.759

Bad thing is, because I've been working

291.28

on this, one, I am not a DIY kind of

293.68

person. I will go figure that stuff out,

295.44

but I'm just I I don't do good jobs with

298.8

stuff. I'm the person that's like not

301.04

likely to, you know, get a nail hammered

303.44

straight in often enough or uh, you

306.08

know, getting an edge exactly, you know,

308

measured correct and all that kind of

309.36

stuff. And so dealing with things like

311.44

this where you're like off by a

313.039

millimeter and you notice it is a little

315.199

bit frustrating for me. So that has been

317.12

a bad thing. But it's going to get

319.84

better because for the next little bit

321.28

I'm going to be doing a podcast instead

322.72

of working on flooring. And I'm going to

325.039

allow you guys to be introduced to

326.96

Michael on the other side.

329.36

>> Hey everyone, my name is Mike Malash.

330.96

I'm one of the co-founders building

332.16

better developers, also known as

333.6

developer. I'm also the founder and

335.6

owner of Envision QA where we help

337.68

businesses take control of their

339.12

software by focusing on what matters and

341.28

that is quality, reliability, and

342.96

support that you can count on. Whether

344.88

you're building something new or trying

346.639

to fix what's broken, we combine custom

348.8

development with thorough testing to

350.8

make sure your systems actually work.

353.28

This way before you get in in front of

355.28

your customers and let them touch it. We

357.12

make sure that it works for them. Uh at

360.4

Envision QA, we don't just build

362

software. We make sure it's solid,

363.52

stable, and ready to support your

364.88

business growth. Check us out at

366.24

envisionqa.com.

368.08

Uh good thing, bad thing. Uh I'll start

370.24

with the bad thing. Um, of course,

372.88

weather is getting colder and I get a

374.88

call from my daughter this morning that

376.88

their HVAC went out. So, yet again,

380.319

another household thing has to be

382.4

replaced that is not a low ticket item.

385.44

Uh, good thing weather is getting

387.28

cooler, trees are starting to change

388.72

color, and we're starting to get into

390.319

fall. So, I'm starting to love it.

393.919

>> It is good. Uh, that is the nice thing

395.6

about the areas that we live in that uh

397.6

you get four seasons. So, you do

399.199

actually get to see like the leaves

400.72

change and stuff like that. Of course,

402.08

that also stirs up allergies, but that's

404.16

another discussion. Today, we're going

406.8

to dive back into we talked about low

408.72

code and no code as our foundational

411.039

stuff and we wanted we touched on AI and

414.56

vibe coding and I think we're going to

415.759

do a little bit more of that uh today

418.639

because we really didn't have the time

420.08

to uh you know give it the proper uh

423.199

attention essentially.

425.52

Now vibe coding uh let's start with the

427.84

idea of vibe coding is that you are

430

working through uh solving a problem and

432.8

you're using AI to generate some code

434.72

and give you and basically allowing AI

437.52

to sort of code as you go and there's

440.88

there are a lot of apps that are I guess

443.36

apps we'll put quotes around those

445.12

there's a lot of things that have been

446.08

built out there a lot of people I've

447.52

seen that are uh essentially putting

449.36

together MVPs minimally viable products

451.599

of some sort using vibe coding Now the

455.919

problem with vibe coding is that there

457.44

is also a huge amount of technology debt

460.479

that ends up involved there because AI

462.88

is not going to have that cohesive

465.599

thought that you need as far as as

467.759

building software. So from a

470.16

foundational point of view I want to

471.52

talk about so you know if if vibe coding

474.319

is in itself maybe not good um then how

477.84

do we find a way to make it good enough

480.879

so that it can help us be more

482.16

productive and that's where I want to

483.919

focus in in my discussion is

487.36

finding is like anything else and this

489.68

goes back to what I've said before um I

492.08

got from somebody else actually is that

494

when you're dealing with AI think about

495.68

it as dealing with a very junior maybe

498.72

mid-level, you know, in this case

500.56

developer. That means you have to be

503.599

very specific.

505.759

So if you're building an app, let's say

508.4

using vibe coding, it can be very

510.96

efficient. A good example is if you

513.519

build a page that is part of your

516.159

application and it g it has a look and a

518.959

feel and a a style and a structure and u

523.36

you know the backend infrastructure and

525.44

architecture is sort of there or

527.44

actually not even sort of is there and

529.519

now you're just building on top of it

531.44

then AI can be very good as long as you

533.839

say given this architecture given this

536.48

style given this look and feel I need

539.04

you to build me a page that does X and

542.48

it's not going to be 100% but it can get

544.64

you very quickly to a lot of the code

547.76

that you normally would be like you know

549.44

maybe copy and pasting or you're like

551.2

just typing the stuff that you've typed

553.36

a thousand times before

555.839

and if it's and the more you can refine

559.279

and define your request the better it's

562.08

going to be. So there's a lot of times

564.08

that I will, you know, if I want to

565.68

build a page very quickly, like a new

567.76

page for an application, I can, I guess

570.16

we'll call it vibe code, but basically

571.68

I'm going to use AI and say here's my

573.839

CSS. Here's a page that's almost exactly

576.399

what I want. Here's the data model.

579.2

These are the fields that I want to

580.56

display. This is how I want to display

582.399

them. You know, this is the general look

584.32

and feel. Then it will spit something

586

back out fairly quickly. you know, maybe

588.24

in minutes it will give me something

590.56

that it would have taken me maybe 20 or

592.72

30 minutes or longer to code out.

594.72

Particularly if it's something complex,

597.04

if it's something where you've got a lot

598.32

of of of lists or a lot of very

600.56

repetitive things where you can say,

603.12

"Oh, hey, I want this page." And it's

605.04

going to have 10 sections in it, it can

607.36

throw up that page very quickly. And

609.2

especially if you just need like that uh

611.6

that skeleton essentially that you're

613.36

going to then flesh out. And that

615.92

actually is the next thing that I would

618.079

say is like keep your request small and

622.16

maybe start with uh particularly in a

624.56

conversation start with an outline.

626.16

Start with something that you can start

627.839

you know a skeleton that you can start

629.6

putting the pieces together. So if you

631.519

think of a page I'm give you an example

633.839

before I pass this on to Michael and his

635.44

thoughts. If you have a page let's start

637.519

with like think of a web application

639.6

you're going to have uh some sort of

641.6

navigation. So, if you know what the

644.16

navigation roughly looks like, you know,

646

maybe you want to do a top menu, you

647.519

want to do a side menu, maybe you want

648.88

to do pop out menus, you want to do a

650.56

menu off the right, the left, the

652.16

bottom, however it is.

654.959

Once you can describe that,

658.72

you can say, "Hey, give me a menu that

661.12

follows these that, you know, follows

663.519

these points that actually hits these

665.839

requirements." And it may not always

667.68

give you all the requirements. It may

669.12

not hit all of them, but it's going to

670.72

give you something now that you can work

672

with. And so what you could do is you

673.68

could take that and maybe it doesn't and

675.279

you say, "Oh, wait a minute. These two

677.12

navigation items need to be a different

679.76

color." And then you can specifically

682.56

drill down into it and say, "Okay, let's

684.88

make these a different color." And this

686.64

is, I know, a very simple example, but

688.959

this is a way that you can actually vibe

690.8

code your way into being more

692.16

productive. You have to start with a

694.8

good foundation. Now, you can use the

697.12

vibe piece to start to put those pieces

699.36

together for the foundation, but if you

701.839

don't want to have a lot of technical

703.12

debt, you're going to have to go back

705.839

and clean it up as you go and make sure

708.72

that it is, you know, what you what you

710.72

are producing is clean and then also

713.44

give it back clean because usually if

715.839

you give it back, if you give it

717.279

something in a certain style that

718.72

follows certain standards, you're going

720.72

to see something that is similar to

723.279

those standards. And then be careful to

725.36

make sure that if it kicks back more

726.959

than you ask for that you don't pull in

729.12

that extra croft basically because a lot

731.92

of times it's going to end up being some

733.92

sort of technical debt. That's like I

736

know that was more of a soap box than an

737.6

intro. Uh but I will step off my soap

739.68

box and then toss that over to you

741.12

Michael your thoughts.

743.04

>> Yeah so typically I look at vibe coding.

747.36

So given my background with testing and

751.279

you know the focus of what I've been

752.959

doing for the last 10 years is more

754.8

test-driven development vibe coding to

757.839

me can be very successful especially if

760.24

you come at it from a testers

762.24

perspective or at least a user story

764.079

perspective. So like you said you got to

766.079

be very descriptive. You have to kind of

768.48

walk AI through what it is that you

771.04

want. The thing is if you actually walk

773.2

it through with clear user stories like

775.519

hey as a user I need to uh log into an

779.2

application. I need a login page and you

781.36

walk through kind of the requirements

783.44

that you need for the page and then AI

786

can generate. Now you can be very kind

788

of clean and straightforward like you

789.279

said like here add a button here and

790.56

things like that. But if you walk

792.24

through it through requirements focus

794.8

through the actual user story, you

798.32

actually can get better clarification

801.2

with AI as far as feedback. The other

804.079

thing I found is as you go through the

806.959

interaction with AI to build the

808.399

application from a user story

809.92

perspective,

811.44

you can actually start out by saying,

812.88

"Hey, write me a test case to do this."

815.839

And it'll build you a test case. And

817.6

then you can kind of then walk from the

819.36

test case into the application like okay

821.44

build me a test case to test a login

823.92

page. Okay now build me a login page for

826.16

this test case now and then you copy

828.959

paste this into your application. What I

831.279

find that's really useful with these

832.959

user stories especially with coding and

834.8

for vibe coding is you can eliminate a

838.72

lot of that boilerplate or at least the

841.04

structural understanding of code. So, if

844.399

you're not a coder and you don't really

845.839

know a language, you can just start out

848.079

with, hey, how do I build a website?

851.04

What do I need to build a website? And

854

it will say, okay, well, you need X, Y,

856.079

and Z. Okay, I don't understand that.

859.279

Ask it. Okay, what how do I set this up?

862

And

863.76

really,

866.16

AI can get you almost there.

869.76

There are times where it will get you

872.16

off the rails. You may need to go find

873.68

another AI engine or just jump out to

875.92

Google and search. Okay, AI tell me I

878.72

need to build a web server. All right,

880.48

it's giving me some steps, but I'm

881.839

having problems. AI is not quite getting

884.079

me there. So, take where you're

885.6

struggling and drop it out into Google.

887.519

Yes, Google still does some AI as well,

889.519

but you might get some better results

891.76

through Google to help you unplug some

894.48

of those issues. So, V coding can get

897.12

you mostly there, especially if you

899.92

don't know anything about coding. But

902.16

getting it to build you something that

904.639

you want or help you understand what you

907.12

want. It is better to go at it from a to

909.92

me go at better from a user story

912.639

perspective and walk it through what you

915.76

want like be very clear because user

918

stories really translate to requirements

920.56

and if you can explain it to AI clearly

923.6

and flush it out the confusion it has I

927.6

think you're going to get your a better

929.519

result of vibe coding with AI than if

931.839

you don't.

933.76

I agree that there I think that's that

935.92

is probably the key is getting the user

937.6

stories right. Um the problem with those

941.279

or I guess the challenge with that when

942.72

you do this is to make sure that you

944.16

also take those to the right uh the

946.32

right level. uh it's very easy uh and

948.88

this is a lot of what we do in

949.92

requirements gathering and I think

951.04

that'll probably be maybe the next

952.639

episode is we'll talk about that a

954.079

little bit

955.759

is we tend to we start with a big story

959.36

usually like I love to use you know the

961.44

ATM example of like I just want an ATM

963.759

where I have an automated teller well

966

cool but then what does that teller do

969.12

what are the features are going to be

970.639

available okay I want to be able to

971.92

deposit money and I want to be able to

973.199

withdraw money okay well is it for one

976.72

account? Is it for multiple accounts? Is

978.88

it how would you do that? Because before

980.72

an AP ATM, it's like, well, nobody had

982.8

an ATM card. So, how would that work?

986.639

What happens? And then there's the all

989.199

the questions around like, what happens

990.639

if I try to withdraw money and there's

992.72

no money in the machine or if I don't

994.88

have enough money in my account or what

996.32

if I try to deposit money and I don't

998.88

actually deposit the cash or I don't

1001.519

deposit the check. How do I validate

1003.519

that check? How do I validate the cash?

1005.199

How do I count the thing? There's

1006.56

there's a lot of these things that goes

1008.48

on and on and on. And so when you're

1010.639

building your stories is

1014.399

it works again great to start from like

1016.56

sort of the high level but then you need

1018.72

to build it into a much more detailed

1021.279

outline. it is and you have to

1025.839

you have to sort of check yourself

1027.76

because I have found too often that

1029.52

people say like and I've had those

1032.16

conversations with customers not AI but

1034.64

just actually

1037.039

human beings human eye um where it's you

1040.959

know saying well this is what I want and

1043.919

as you have a conversation you realize

1046.64

that what they said they want is not

1048.96

actually what they want um there's a lot

1051.76

lot of people will hear things like uh

1054.559

CRM or ERP or uh EMR or all these other

1060.32

like tools and systems that are out

1063.2

there and they'll say like so and so

1064.64

said I need this I need an HR system and

1069.2

you find out that it's a somebody that

1071.039

somebody that has zero employees and

1074.4

like uh I don't know if you really need

1076.559

a HR system. What do you actually what

1079.52

problem are you actually trying to

1081.28

solve? And I think too often we start

1083.36

with the story instead of the problem. I

1086.799

think if you take it back a step and say

1089.44

this is the problem. This is the

1092.4

description of what I think the solution

1094.16

looks like then a lot of times then

1097.2

you'll be able to get some stuff back

1098.64

that says well you'll maybe AI will ask

1100.64

you question say well what about this or

1102.16

what about that or it may say that

1104.32

sounds like this kind of a solution.

1107.12

Should I look up some more about that?

1108.88

Should I educate you on that? And those

1112.32

are the ways that I think that you can

1113.919

step into uh AI and vibe coding and help

1118.48

yourself get to a solution that is

1120.88

actually a little more useful with a

1122.88

little less uh overhead and some of the

1125.52

extra stuff that you might have to clean

1127.2

up after the fact.

1130.32

I think we will I'll give you one more

1132

because we're we don't want to go too

1133.6

long because we're going to actually

1135.2

spoiler alert we're going to spill into

1136.799

some of this in the next episode.

1139.12

Thoughts?

1139.52

>> So I I kind of want to touch on the

1141.039

negatives of this a little bit. So if

1143.44

you're going into AI and you you're

1145.52

asking those things or you you have a

1148.559

problem but you haven't really defined

1150

the problem, be careful with AI vibe

1154.08

coding or even asking AI for too much

1157.039

input into what you need really to go

1159.52

talk to your customer about because you

1161.76

could get what's called confirmation

1163.76

bias. You could essentially go down a

1166

track of AI is telling you what you

1169.36

think you need and you're basically

1171.28

feeding AI. Oh, okay. Tell me more about

1173.76

and you're going down the wrong rabbit

1175.2

hole and you're actually going away from

1177.36

the problem or you're going away from

1178.72

the correct requirement. So, make sure

1180.799

that you do kind of check yourself as

1183.6

you're going through this and even check

1185.44

with the customer to make sure that

1186.799

you've got the right requirement before

1188.559

you go too far down building a system.

1191.919

That's a good Yeah, it's uh when in

1195.36

doubt, check yourself. Particularly when

1196.72

you're in an area where you don't um you

1199.52

don't have domain knowledge, where

1200.799

you're not really sure what you're

1202.16

doing. There's too often uh this goes

1205.039

back to and you then we'll wrap this one

1206.72

up. I think the thought is like there's

1208.32

been a lot of times that we've had

1209.919

customers, we've complained about this

1211.36

where or potential customers, prospects

1213.039

that have said, "Hey, I want to build um

1217.52

their project definition. I want to

1218.96

build eBay for uh honey or something

1222.16

like that or I want to be the Amazon for

1225.679

uh flower delivery or something like

1227.919

that and that's it. That's like their

1230.159

story. That's not going to cut it. Those

1232.48

kinds of things like you need to really

1234.24

spend some time researching and thinking

1236.24

about what it is you're getting into.

1238.24

And a lot of times I think you really

1239.919

need to go look when you get those

1242.4

answers. Do not take those answers

1244.08

because AI is still based on the

1246.08

internet basically. Don't take those

1247.679

answers as gospel because there's a lot

1249.52

of times you're going to get an answer

1251.12

and when you really start looking at it

1253.679

um you're going to find out that it's

1254.88

wrong. Uh I don't know how many times

1256.64

I've I've corrected AI and it's like oh

1258.799

yeah you're right. I didn't think about

1260.24

that or I wasn't using that or I was

1263.44

thinking this or you know I was focusing

1265.44

on this other thing sometimes just like

1267.84

human beings.

1269.6

But the direct is always going to be the

1271.84

best way. things like send us an email

1273.52

at infoddeveloper.com and let us know

1275.679

your thoughts. What do you what are your

1278

thoughts? I I would love to hear like

1280

low code no code vibe coding stories. Uh

1284.32

I know a lot of people are big on I see

1285.76

a lot of people that are projects out

1288.159

there that are potential customers like

1289.679

we want to use this, we want to no code

1291.2

this and low code that and vibe code

1292.88

that other thing. And I'd like to see

1295.44

some people that are like particularly

1296.799

that have been successful that have

1298.32

taken one of those systems completely to

1301.039

production and served, you know, more

1303.76

than five people at a time or something

1305.12

like that. And I've seen some, I'm not

1307.2

saying they don't exist. I've built a

1308.96

couple of things that are in these

1310.64

realms that have done that. But I would

1312.799

love to hear some other stories because

1314.4

it seems like they are few and far

1316.08

between. Most of them seem like they are

1318.48

things where people are like, "Okay, we

1319.84

built this thing in low no code. We vibe

1321.919

coded it. now we have to rewrite it

1323.6

completely in something else because

1325.2

they really didn't get they just got a

1327.28

like a play solution and not something

1330

that was production ready which is often

1332.96

the case with uh code generation tools.

1335.039

I digress again but uh check us out

1338.799

developer.com you can leave us comments

1340.64

there. We've got plenty and plenty of

1342.96

content there. The YouTube channel

1344.72

developer YouTube channel we've got all

1346.88

of like multiple seasons of episodes not

1349.679

to mention a lot of other stuff out

1351.039

there.

1352.159

Um, you can check us out on X. We are at

1354.96

Developer. We have a Facebook page.

1356.88

We're all of those good things. So,

1358.799

wherever you wherever you subscribe to

1360.64

podcast, leave us a review, leave us

1362.159

some feedback. We would love to hear it

1364.24

and uh help you out wherever we can with

1366.559

new, you know, new seasons, new

1368.24

episodes, new topics that are the ones

1371.039

that you requested because we're here

1372.96

for you. That being said, I have said

1375.52

too much. Go out there and have yourself

1377.36

a great day, a great week, and we will

1379.679

talk to you next time.

1383.36

Uh, bonus materials. You got to see the

1386.559

genesis of this this episode. So, we're

1389.28

going to keep this one a little bit

1390.32

short because

1392.559

you know what? You can't always have

1394.159

bonus. It's not Christmas every day. You

1395.76

don't get bonus after every episode. And

1397.6

depending on how this next one goes, you

1399.2

may get a lot of bonus stuff. Thank you

1401.76

so much for your time. We do appreciate

1403.36

you. Appreciate you guys hanging out

1404.96

here. uh all the ways I said give us the

1408

feedback you know it's all the feels

1410.159

it's all the feeds all the feedbacks uh

1412.64

let us know and we would be love love

1414.72

love to hear from you if you want to be

1417.039

a you know on the show let us know uh

1419.6

we're open to discussions with a lot of

1421.76

different people for our conversations

1423.6

our interviews and uh we're more than

1426.08

happy to bring people of all stripes

1427.919

because that is what makes us all better

1431.039

go out and have yourself a great Today,