🚀 Add to Chrome – It’s Free - YouTube Summarizer
Category: Entrepreneurship
Tags: AIDevelopmentEntrepreneurshipNetworkingSaaS
Entities: Ben WebbCursorGitHubMake.comMicrosoftRenderRoboFlowSuperbase
00:00
If you have ever wanted to build your own SAS but felt like you weren't qualified or didn't have the skills to do so, this video will completely change your mind on that because we have Ben Webb coming on here who until recently was completely non-technical and since then has been able to make over $100,000 with this very, very interesting dental
00:15
sass that he's built. And this is an awesome example of people coming to the AI agency route, finding a great use case, building a software, and in Ben's case, feeling like he's completely out of his depth, but pushing through anyway and building an incredible business as a [music] result.
In this video, Ben breaks down his entire process of how he found his idea, how he validated it, the
00:32
exact tech stack that he uses that costs less than $100 per [music] month to be able to build his own SAS and walks you through the process of building the front end, the back end, and so on [music] as a complete beginner. So, this is a really, really valuable one for anyone looking to build any kind of SAS.
Hope you guys enjoy this one. All right, Ben, great to meet you, man.
Thank you for coming on. Um, you've had a pretty
00:49
awesome run here with your uh vibe coded SAS in the dental industry. >> Yeah, thank you for having me on, Lamb.
Back on in the flesh this time. I'm really excited for this one.
That's going to be a banger. So, what I've got for you guys today, as you can see on screen, is a five-step, a simple and actionable framework that you can take
01:04
to go from idea to a SAS MVP, even if you're someone someone like myself who's never touched code before, doesn't have a developer or a massive budget. Moving into step one of the framework, I actually want to start with my story here to really give you guys an understanding of uh what a realistic uh
01:20
experience and expectations of this was cuz mine certainly was not a linear story. So random Bulgarian dentists almost paying Microsoft $1.5 million.
So basically I met this gentleman um on the other side of the world in Eastern Europe just about as far as far as you can get from Melbourne came in through
01:37
my school group and we kicked it off. We had a common interest in AI and automation and after talking for a while we had this idea to train an AI X-ray diagnostics model.
was something that he wanted for his own clinics actually as a diagnostic tool and I liked the idea of it for a SAS um but I at the time had no
01:54
idea how I was actually going to leverage it um because the AI diagnostic side of of dentistry is dominated by some massive um corporate companies that I have no business competing with. Now, we gave this idea some serious consideration and we actually went to Microsoft as yours team and got a
02:10
shockingly a $1.5 million quote to train this model, which was slightly more than I was expecting. And to be completely honest with you, I actually considered uh cutting out both of my kidneys and selling them to make this work.
But uh moments before grabbing a scalpel, we
02:25
actually had a bit of an epiphany that we'd already done the hardest part, which was because he had been a dentist for many years running his own clinics, we actually had all of the data we needed to train the model. He had thousands and thousands of X-rays sitting around basically gathering virtual dust.
So we hired a dev team and
02:42
got it done in 45 days, which was far far less than I was actually expecting. >> So that's just using the pairs of like input and output.
Like this is the X-ray and this is the human like human diagnostic off the back of it. >> Yeah.
Exactly. So we use a platform called I'll drop the source here called RoboFlow.
So this is basically a um
02:59
computer vision machine learning kind of thing >> training platform. >> Yeah.
Yeah. So I just want to I just want to say quickly here this is a bit of a unrealistic situation where I've been presented a unbelievable resource basically from the sky.
So most SAS will
03:15
not require something like this to get started up. With that being said, if you put yourself out there, and I know you can speak to lamb, uh, you can speak to this lamb.
If you put yourself out there and you network with people and you post content, things like this are just going to happen. And that's going to be the spark to starting your business.
You're going to meet your business partner. You're going to get an asset like this.
03:31
So, how to actually find a problem. Now, generally, the way I like to look at this is inside any industry, there's a pool of problems that exist.
And then when a new model or AI software comes out, it sort of unlocks a few of these problems to actually be solved. Now, two
03:47
concrete examples here to really illustrate this to you guys. Voice realism and latency technology getting finally good enough with things like 11 Labs or GPT real time.
And all of a sudden, you've got, you know, customer service, inbound calls, outbound calls, um, reactivation, receptionist, this massive chunk across hundreds, if not
04:04
thousands of verticals basically opens up overnight. Same thing with text, image, and video generation models.
Sora Nano Banana content creation which is you know this massive chunk again across hundreds of verticals opens up overnight. So I don't want to hear anyone in the comments saying you know finding a problem's too hard where do I
04:20
start? Like this is the market we're in right now.
This stuff is happening on a weekly basis. Okay guys very quickly.
If you're an aspiring entrepreneur and want to start your own AI business and you haven't already joined my free school community it's down there in one of the links in the description below. Has my full free course on how to start your own AI agency as a complete beginner.
04:36
and you're surrounded by over a quarter million people who are also striving towards the same thing. There's no better place on the planet right now to be surrounded by like-minded people and you get free weekly Q&A with me where you can ask questions directly to me about how to start and scale your business.
I'll see you in there. I literally was just talking with Corbyn about this.
It's I call it I call it
04:52
spawn trapping like like you're basically spawn camping. The new model comes out there's like a fertile plane appears is what I call like a step change in the model capabilities.
Okay, the image models are now way better and you can put text in it. I remember that was like the GPT40 image gen came out
05:07
earlier this year and suddenly you could put text in it and it had like much much better image generation capabilities. That's a step change and there's all these new business opportunities that get unlocked by it.
So this is a you can literally just camp the new models as they come out and wait for a wait for a really big step change to come.
05:23
>> Yeah. And it's not like it's hard to see what's coming next, right?
If you pay even the slightest bit of attention, you can see what the next move is. So also something to be cognizant of here.
Crumbs can be worth an 8 to nine figure valuation. So what I mean by this and I'll pull from my personal experience here when I was going into dentistry the
05:38
problem that I'm solving with my SAS which is um manual treatment plan creation basically is not problem 1 2 3 or five in the industry but because dentistry has a large enough total addressable market that problem is still worth pursuing. It's the crumbs of the industry if you like but it still can be
05:55
worth an 8 to9 figure valuation. So when you're looking at an industry, don't be surprised when the, you know, top 10 biggest pain points are already solved by companies that have got there faster than you.
But it's still worth, if it's the right market, pursuing the other problems in there as well. Now, also the
06:10
world's completely changed because what a lot of people their attention is drawn toward when they're thinking about this is okay, the barriers to entry have been lowered from the coding side, but they've also been lowered drastically from theformational side. And again, pulling from my personal experience here, I'm not a dentist.
My experience
06:27
as far as dentistry goes is I've got a few teeth and I've been to the dentist a few times and had them poke around in my mouth. Yet somehow I have been able to build a SAS around dentists.
Now you can leverage tools like uh Chat GBT's deep research for example to pull from
06:42
thousands of data points on the web and then present all of the information to you in a concise articulate manner where you actually have all the information you need to understand the the workflows the preferences of the professional that you're building your SAS. So how to actually go about finding a problem?
06:58
Well Chad GPT is a great place to start but there truly is no substitute for talking to profession like real people in the industry. So friends, families, colleagues, anyone you know who is a dentist, a chiropractor.
And generally inside um the problems that you're going to be looking for will be inside two
07:14
pillars. How do you get customers and how do you fulfill the service?
Now of course there smaller things under these two umbrellas. You know customer service, onboarding, lead qualification, etc.
But these are the two sort of pillars that you want to start with and simply ask them what makes you want to punch your computer? What makes you want to throw your computer in the trash?
07:29
Like this guy here. Um there is always something.
There is no industry that exists, not on earth, that does not have some things on a daily basis that just annoy the heck out of the owners. In fact, I first want to ask you, Lamb, this is obviously one of the most crucial steps to do properly, otherwise your SAS will crash and burn.
So, I want
07:46
to ask you, is there any pitfalls that you commonly see beginners doing here? >> It's always too big.
Like, they're trying to solve way too much. And I think they're underestimating the difficulty of the dev and overestimating their ability to to vibe code.
um at at
08:01
the end of the day like you like start far far far far smaller than than you think and and like uh like I've said before um previously is like you want to make sure that if you're picking and a problem to solve make sure that the the gold mine is built on top of the gold,
08:18
right? Like there's certain use cases, there's certain uses of these models that instantly create value in a way that's like so clear and undeniable that it's it'll be a hit right off off the rip.
And those are the ones that you should of course be trying to find. And there's other software use cases where you can you know like task management or
08:35
like productivity or like like maybe a little bit of scheduling here and there. In the case of image generation models, video generation models, uh computer vision in your case, uh there are some very very very immediately clear value creation moments you can use with the you can get with these models and the
08:50
rookie will kind of shoot for something they think sounds interesting and build this fancy software and try to iterate until you get product market fit. um when you could probably just have like really like cracked it off the off the tea um by just building over the over the go over the gold in the first place.
Now, a bit of a pro tip here. Once you
09:07
have found an industry professional, this is going to be an extremely and I cannot stress this enough, an extremely high lever contact to have. So, nurture this relationship, keep in touch with them because you're going to need to use them later.
So, that takes me to step number two, which is building the
09:22
output. Now, I want to put an emphasis on what I said here, which is speed is key.
What you're not doing here is building a dashboard, a UI, a login system. You just need to build the core output that solves the problem.
Okay? Because what you need here before you go spend time and money building something
09:38
else is fast feedback. So, realistically, you have two options to do this.
So just just to just to clarify that you're meaning the core functionality of the app in your case the model that's able to take in an X-ray and and provide the like JSON output to uh give it the diagnostic
09:54
right. >> Yeah absolutely.
So for me it was more the actual report like can we is the uh the treatment report that we generate for the patients does that actually suffice. So yeah the the AI model was the hard thing to get but um yeah definitely just the core output that solves the problem and not the fluff
10:10
surrounding it. So, two options to do this, make or n.
Now, if you're not familiar with these platforms, YouTube, Liam's channel in general has a lot of content about how to learn these. Um, but generally, this is the approach you want to go for if your output requires heavy logic or complicated processing to
10:25
be done. Now, bit of a hack to speed up this process is getting the blueprint generated from chat GPT.
Now, make and ed have got I think they're starting to see some vibe automation, if that's what you want to call it, where you just prompt inside these platforms to build
10:41
it. Um, I personally don't think they're all the way there yet.
So, this is this is just what I use to do the heavy lifting for me. Um, but that is actually inside of chat GPT.
So, I said generate an importable make.com JSON blueprint for a scenario using the following instructions. And then you're obviously going to um have a comprehensive prompt
10:58
there about what you want that blueprint to to do and look like. Then it'll give you the JSON file back.
So download it to your desktop and then when you're inside this platform whether it's make or or nadm import. So hit the three dots import the import the blueprint and then it will populate all of the modules for you.
Now this won't be perfect out of
11:14
the box. You will have to do some tweaking but it definitely will do 90% of the heavy lifting for you.
Option number two is vibe coding. And now as I said I'm going to get into a lot of depth about vibe coding tech stack all of those things.
But to put it simply right now, this will be your approach if you have like if you can't no code the
11:30
output if it's a little more complicated or requires heavy processing. Um this will be your approach.
Now the balance you want to strike here kind of goes to what I was saying before is there's sort of two ends of the spectrum when you're looking at this. This end of the spectrum is you don't spend enough time developing the output and there isn't
11:46
enough substance there to get proper feedback on it. The other side, and this is what beginners tend to gravitate towards, is overdoing it, spending too much time, and you obviously want to find a nice balance there.
Have just enough substance that when you go talk to a professional to get market validation, they're giving you proper and useful feedback. So, a bit of a rule
12:02
of thumb, if your output is useful but ugly, perfect. If it looks good and it's pretty, but it's [ __ ] useless, you have wasted your time.
Okay. Now, my first solution here, um, this is giving me stroke looking at this.
uh uploads an X-ray, gets the AI detections from the
12:18
models, processes the conditions, treatments, and urgency with um GPT steps, and then creates and sends the PDF treatment report to the patient. Now, funny story about this.
Um I had just sold my marketing agency at the time. We just got the AI model trained, and I was super pumped and excited to
12:34
start working on this shiny object syndrome in the worst way. I basically put in two backto-back 16-hour days building this, working myself to death, and on the end of the final day, I actually came down with a fever.
Now, for anyone who's had a fever dream, um, you'll understand how awful this is, but I was basically in here, you know,
12:51
terrified of make.com, you know, errors and broken JSON and disconnected modules and stuff. Um, it was, uh, yeah, any any founders's worst nightmare, basically.
So, uh I guess the moral of the story here is take it one step at a time. Only build what you need to and um get your 9
13:07
hours. >> But it is this just goes to show like when you find something that you're really interested in and you know like you can put in the you can put in the work and I think that's what a lot of people are like lowkey craving is something that they're so excited for and has so much potential that they they
13:23
bang out 16our days and they and they're probably still smiling at the end of it. >> Yeah.
Exactly. And for me it was like I can build a medical grade dental sass like what like and I just jumped straight into it and worked my ass off.
So yeah, you're exactly right. >> Your learning of of make.com that was just YouTube tutorials and and tooling
13:39
around like that and then solving some of your own problems. >> Yeah, pretty much.
Um to be honest, I got most of my experience like I kind of learned in the process. I'd tinkered with voice agents, done some small things like that before, but most of my learning from make actually came from having a crack at that and and building
13:54
it out. I was kind of learning as I was going along to be honest.
So now is time to validate the product. So this means um get out of your cave, get out of your dungeon, go in and talk to some professionals.
Okay, validation does not happen on chat GPT. Now this is where what I said before about having that um
14:10
industry friend. Keeping in touch with them, nurture that relationship because here and it's going to continue to be super useful.
Now most beginners think that validation looks like this. This is what I like to call the Greg approach.
Now, Greg, he spends two weeks making a pitch deck. He validates his SAS with
14:26
chat GPT. He's always in the mindset of adding one more feature before actually talking to anyone, and he takes ages to get any useful feedback.
Okay? You don't want to be Greg.
Now, this is how winners do it. And this is with the chat approach.
Now, I know it's a it's a bit abstract, but this is what the kids are doing these days, so uh I got to ride
14:41
the wave here. But Chad, he got super super clear on the industry pain points that his SAS actually solves.
He had just enough substance to get proper useful feedback. He spent the morning preparing, afternoon doing, cold called cold called cold called cold called cold called cold called cold called cold called cold called cold called cold called cold called cold called cold called cold called cold called cold called cold called cold called cold called cold cold walked in Chad was oozing charisma now personally I took
14:57
the chat approach not because I am a Chad unfortunately I'm not as handsome as this gentleman here but because in the past I have taken the Greg approach now Liam I'm sure you're the same most people are when they're starting their first businesses they take the Greg approach so you guys can be you guys can be wise and learn from the mistakes of
15:12
others instead of starting with the Greg approach okay so here I've actually I started with cold calling my dentist and to be honest that was such an awkward Ward conversation trying to explain to the receptionist. I wanted to make an appointment, but I didn't want to get my teeth cleaned.
I wanted to talk to the
15:28
owner about my new software startup. Took a while to her to get the memo about that, but uh friend of a friend as well, dad's college mates, like literally any professionals, um I could, you know, get in touch with and set up meetings.
>> What was the promise? Was it like, hey, you're not really going to get anything out of this, but can I just talk to you
15:44
or like how did you frame that? >> Yeah, pretty much.
So I I know it's a bit of a scary thing to do, but I genuinely think that if you call people and say something to the extent of, "Hey, I'm building a software product. I would love 5 to 10 minutes of your time to get some feedback on if my product
16:00
solves X problem for your industry." Now, the success of this actually working and setting up a meeting or just getting treated like every other cold caller does and being told to basically get lost is going to be is your problem actually something that resonates with this person? because you can have all of
16:16
the Chad charisma that you want and all of the sort of going with the approach of market feedback and testing things and blah blah blah, but if that's not a problem that actually resonates with them, they're going to, as I said, treat you like every other cold caller. >> So, you're kind of like subtly flattering them, but also like peing a
16:32
curiosity around like this person's called me and said that he's interested in just like hearing my thoughts and from my position of expertise on this thing. And it's also appears to be something that I struggle with right now.
So maybe [ __ ] like what's the worst that can happen? I take this call.
16:49
>> Exactly. If you're a business owner who's interested in what genative AI can do for your business, you can get in touch with me and my team at Morningai in one of the links in the description below and we can start your entire AI transformation process going all the way from the education and training of your staff to the identification of the best AI use cases for your company all the
17:05
way through to development and beyond. We've worked with some of the world's biggest sports teams and also publicly traded companies.
So rest assured you are in good hands. >> Now for me at least in dentistry and you could probably extrapolate this out into your own market in dentistry since the beginning of the technological age in
17:21
like actually implemented in dentistry the technical and the software aspects have been dominated by these prehistoric slowmoving corporate giants. So for someone like a young person, you know, who speaks well, well presented, for them to come in and kind of have a bit of an an innovative approach, um that's
17:36
another thing that's actually going to be very attractive to most um owners in most industries anyway. So another thing to keep in mind there.
Now, if you do get desperate, you don't have enough warm network professionals to get by, literally Google search, dentists, chiropractors, whoever near me, cold, call them, set up the meeting. Right?
I
17:53
know it's a little scary. the people the archetype is going to be drawn toward starting a SAS probably isn't going to be too excited about doing this but uh it's what it takes.
So step four is time to build the front end. So now you've actually validated that your output solves the front end.
You're going to build the front end. Now for those of
18:08
you who don't know the front end is basically the interface that users interact with buttons, screens, forms, things of that nature. So we're going to head into our old friend chat GPT or your preferred LLM to do some brainstorming.
So, we need to figure out the specific UI sections that you
18:24
actually need. So, I'm talking um do you need a upload, a processing, a results, a dashboard, a settings, all of those things.
You need to get clear on what components you need. And then you need to create a lovable UI generation prompt.
Now, I know you're an enthusiast of uh Lovable, Liam. Is that right?
18:40
>> I've used Lovable and Bolt here and there and like we've done a course on on Bolt. Um but because it's my team actually uses a lot more than me.
Um, we do it for like rapid MVPs to show clients like on when we're doing our inerson audits, we'll build literally build a prototype right in in front of them and just show them what it would
18:56
look like. So, we use it more for like visualizations, but I'm definitely more in the uh in the cloud code and and uh and VS Code sort of arena now.
>> Gotcha. Yeah.
Well, it's an awesome platform. For those who don't know, it basically vibe codes uh VI codes the front end for you, the UI.
So once you've got clear on the actual
19:12
interfaces that you need, you're going to get chat GBT to create a lovable UI generation prompt for my SAS dashboard, my settings, my whatever it is. So now you're going to have a separate prompt for each UI for me.
As I said, upload an X-ray AI results report view and send dashboard. Pretty simple stuff.
Now
19:28
you're going to paste all of these individual prompts into Lovable. And bit of a pro tip here.
You get five free credits per day with Lovable. Meaning you can generate five UA screens for free in one day.
If you were to tell any developer that 5 years ago, they'd lose
19:45
their mind. I know I sound like a broken record here, but it's truly insane stuff.
Now, once you've got those front ends generated, click the code button, click download, and you will now have the downloaded code for each front-end component on your computer. Now, quick pro tip here.
If it gives you something slightly off, don't worry about it. You can fix it later.
things, spacing,
20:01
alignment, colors, anything that's drastically wrong, obviously regenerate. So, this was my UI here.
Um, nothing too crazy. So, that is the AI detections page.
And I generated this um ages ago when lovable was a lot worse basically. So, this is okay, but you guys could
20:18
definitely generate something much much better. Uh, now moving on to step number five, the fun stuff.
So, you have two things now. You have your output from step two, the back end, kind of parts of the back end, and the front end.
So, it's time to stitch these two components together to vibe code. Now, before we
20:34
get into the vibe coding juicy stuff, I do want to talk about the Dunning Krueger effect because coding, the experience of coding hasn't really changed, but the emotional roller coaster of it has been compressed into higher highs, lower lows, and into a shorter time period. So, for those of
20:50
you familiar with the Dunning Crruder effect, you got confidence on one scale and competence on the other. So when you start vibe coding, you actually you prompt and then you deploy it and you see the feature has been added.
You are going to instantly scale mount stupid peak confidence. Okay?
You're going to
21:05
be telling your homie, "Bro, I'm basically a full stack engineer now." But then something is going to break. It's called the value of despair.
Probably going to be looking like this right here. And you might think to yourself, "This SAS thing isn't for me.
This is too hard." Then if you thug it out and you stick through it and you
21:21
have a little bit of persistence, you start to get a little bit more comfortable. You'll fix some bugs, understand the folder structure, and you will slowly start climbing here.
Increase confidence, increase confidence until you arrive at this point here. Curs is going to feel like your co-founder, and all of the pieces are going to start falling into place.
Now,
21:38
I know it seems like a bit of a weird thing to take the time to explain this to you guys, but it is very important for you to understand the emotional journey that is ahead because if you don't know and you aren't optimistic that things will eventually work out, you are likely going to quit here and end up becoming this guy. So just
21:53
understand that. Now, vibe coding for beginners, the juicy stuff you've all been waiting for.
So this is the tech stack that we're going to use. Now, as I go through these platforms, there are alternatives for each of them.
I'm just going to give you the recommendation based on my own personal experience here. So first one here is cursor.
Now
22:09
what cursor is, as I said before, is the interface on your computer that allows you to leverage these different models to actually edit the coding files on your computer. Now this is not the model.
Um the models you will still be able to select, you know, anthropic or open AAI's models. Now, two quick tips here.
Number one is keep an eye out for
22:25
new model promotion week. So, what's happened over the last month or so as Opus 4.5 and GPT 5.1 codecs have come out, they actually let you use these new models for free.
So, this isn't just a one-off thing that happens rarely. Like 60 to 70% of the time you are going to be using the new, which of course is the
22:42
best most competent models completely for free. So, cursor is already cheap, but this makes it that much cheaper.
Also, use the dictate feature, this one here, for faster prompting. Okay, moving on to GitHub.
Now, GitHub, you can kind of think of it as the storage for your code. So, you write, edit the code with
22:57
cursor on your computer and then you send it to GitHub to be stored. Okay, two pro tips here.
Create a private repository on day one and connect it to your cursor project. Now, that connection is going to be made much easier if you do this step here, which is before you create your cursor account, create your GitHub account and
23:14
then when you're creating this one, sign in with GitHub. That'll make the connection so much e easier and save you a lot of headaches.
Yeah, GitHub can be pretty confusing for the for the beginner and it it took me a very long time to properly wrap my head around like all the different things uh in
23:29
there. So, uh I would definitely uh make sure you've got that signed in.
And chat between your best friend and understanding what each of the different comm command means um because it can get can get very difficult for a non I'd say this is one of the more difficult parts of getting into it's like understanding
23:45
how to how the the terminal works, how like you're installing packages and things like this. It's it's al it's more so like the environment of setting up uh your computer in the coding environment properly which I'm sure cursor helps out with a lot now but um that has traditionally been like a very sticky point for beginners getting into
24:00
alongside GitHub. >> Yeah, absolutely.
And if you're struggling there just know you're definitely in the valley of despair. Just push through.
So number three is render. So this is where you actually deploy your code.
This is actually the sort of makes it visitable by users. Gives them a URL to interact with.
And my pro tip here is the backend logs will
24:17
be your best friend for troubleshooting. So you can copy and paste them into cursor.
I'm actually going to give you guys a hack on how to do this automatically um in a second. And then finally is uh Superbase.
So Superbase is a data storage platform. So it will store user information, login details,
24:32
things of that nature and it will talk to your back end. Now you also don't have to do anything in Superbase.
So because cursor knows the exact storage structure that you need in this platform. You can ask it and some models will do this automatically uh to generate an SQL uh file for you.
SQL is
24:48
just a data language. You can then paste that in superbase and it is basically like a prompt to superbase.
So it will generate all the tables and columns and stuff that you need for you. Makes things super super easy.
So this is the tech stack. Now as I said this is this is the perceived barrier to entry.
The
25:04
barriers to entry have slowly slowly been dropping. They've now fallen off a cliff and this is kind of what remains.
people like okay I want to build a SAS I want to get an application what do I actually need I don't understand the components or or what goes into it so I promise if you guys just create a account with these four platforms you
25:20
connect all four of them that is basically as technically challenging as it's going to get for the most part there's obviously going to be hiccups along the way but getting started and getting over this original hump is where the biggest headache likely comes from I've just sold that for you now this is all less than 100 bucks a month I'm not
25:36
going to go through the indiv individual prices of these but like this is insanely cheap. And again, there's things like the promotion models that make this even cheaper.
So, all of that to say, uh there's really no excuses. It is unbelievably accessible.
And then finally, for my pro tip on fixing bugs.
25:51
So, normally the process for fixing this is you go into render your backend logs, copy and paste the errors um from the logs here, and then paste them into cursor and say, "Fix this." Now, what you can actually do is use render API to fetch the latest logs for you. So, you don't have to copy and paste them.
save
26:07
those logs somewhere in a file and basically tell cursor, hey, when this script runs, um, read the error and fix the issue. Now, for those of you who are beginners, not familiar with this stuff, probably sounds like I just spoke a foreign language to you, but my recommendation would just be screenshot this and then when you come up to this point and you're confused, uh, put this
26:23
in chat JPT or something, get it to explain this to you and run you through how to set it up. So, that takes me back to where we started, which was the big bang of vibe coding.
So again, this the barriers to entry have been slowly slowly slowly dropping and they've just
26:39
fallen off a cliff. This is the first time that a complete beginner with no coding experience, not much money, no developer can make an impact.
You can build something significant. So I encourage all of you to don't be Greg.
Take the chat approach and really apply
26:54
yourself and go and build something incredible. >> Yep.
Couldn't agree more, man. And thank you for breaking that down.
I think that's going to demystify a lot of the process for the for the beginner and you've communicated it really clearly for uh for people as well. So if you guys were skeptical before um what's that that quote that I really like
27:10
recently uh pessimists sound smart optimists make money and uh this is truly one of those eras where the pessimist about vibe coding and about like being able to build software as a nontechnical person. You may sound smart now but there's absolutely no way that
27:25
you're going to make any money off of that that position. And there's people like yourself out there, and like I said, many, many, many more who are following the same path and going, "Fuck it." I back myself to be intelligent enough to figure this out.
And I tell you, if you're watching this channel, you probably are, cuz uh I've I've met
27:41
so many of you people in person now. I've met people here through these podcasts, and I'm always blown away by the the quality of people who are watching these videos.
So, if you're watching this, take uh permission from me and permission from Ben to [ __ ] shoot on this stuff next year, guys. Like there is you will regret this to
27:58
the freaking day that you die if you don't do something about this now. And it's the decisions that are you're probably like stressed about it and stuck on the fence.
If you can just make a decision and say I promise you like it is going to be easier than you think. Like for for example for me getting back into the engineering recently getting into learning this like a gentech
28:14
engineering workflow and claude code and how to like it felt out of depth even for me and within literally like two days I'm like oh okay I kind of get this and you start getting the like essentially the the dopamine of like the the small wins and once you're getting small wins you are on the way and the
28:32
way that uh Ben's laid it out here is exactly how you should be approaching it as a beginner. So really appreciate your time, man.
And uh I'm so happy to to hear about your success and I'm sure you're going to go from strength to strength. Um is there anything else you'd like to to wrap up with then?
Um how can people get in touch with you if they want to? >> Yeah, so school.comarketingmastery.
28:50
So I've got um vibe coding stuff um AI lead gen models, ton of ton of just AI general content in there. So um if you're interested in connecting with me, um shoot me a message on there.
>> Sweet, Ben, mate. It's been great to meet you.
>> Pleasure. >> Congrats on your success and uh hopefully I'll be back on here soon.
>> Yeah. Happy new year to you, you land.
29:06
Have a good one. >> Likewise.
So, that is all for this episode of the podcast, guys. If you want to see something similar that I really think you'd like, you can click up here to watch another one.
And remember, if you think you have a story worth telling and some valuable insight you can share with the community, you can fill out my podcast application form in the description below. I'd love to have a chat with you and get some
29:21
exposure for your business. Aside from that, guys, that's all for the video.
Thank you so much for watching and I'll see you in the next