I Asked GPT-6 Astra To Build Me A $437/day Side Hustle

0:00If I suddenly went broke tomorrow and

0:01you stripped away my audience and my

0:03clients, this is exactly how I would

0:05make money on the new GPT6 Astra. In

0:08today's video, I'm going to show you

0:09exactly how to find an idea, how to work

0:12out how to sell it, and how you can use

0:14GPT to execute the entire process. I

0:16truly believe that Chat GPT6 is one of

0:19the first AI models that could actually

0:21make you a millionaire from scratch.

0:23Because it's not only intelligent, like

0:24a Claude Fable 5.1, but it's also really

0:27good at doing stuff. It's great at

0:28visuals. It's great at design. It's

0:30great at agentic flows. It's great at

0:32computer use, which pretty much give it

0:33all the tools to start and run a

0:36business. Today, we're going to cover

0:37three business models: AI consulting,

0:39digital products, and a physical

0:41product. And I'm also going to show you

0:42the system so you can work out the best

0:45business to start for your specific

0:47situation. Because GPT isn't just good

0:49at doing stuff, it's also a fantastic

0:51planner and a great brainstorming

0:53partner. All right. So before we build

0:54three business models from scratch with

0:57chatbt6, the first step is that you need

1:00to identify an opportunity. So for the

1:01sake of today's video, I'm going to tell

1:03GBT to disregard my audience, my moat,

1:05my distribution because I understand

1:07that the majority of you don't have many

1:09clients to market to. And that's okay

1:11because I want to show you how I would

1:12genuinely go about doing this from

1:14scratch. So the prompt I use and all the

1:16prompts will be available in my school

1:17community for free below is help me

1:19identify a business I can realistically

1:20start and execute. Act as a rigorous

1:22business consultant, practical,

1:24commercially minded, and willing to

1:25challenge my assumptions. Your objective

1:27is to find the best fit between my

1:29capabilities, circumstances, and a

1:31problem that customers will pay to

1:33solve. Ultimately, this is business in a

1:35nutshell. You need to have product

1:36market fit. So, you need a product that

1:38customers actually want. And you need

1:39the skill, expertise, or the systems,

1:41which I'll teach you today, to deliver

1:43that product in an efficient way to your

1:45customers. For this exercise, assume I

1:47retain my skills and experience, but

1:49start without my audience, start without

1:50my customer list, or significant

1:52capital. Stage one, diagnose. So, we're

1:54going to get GPT to diagnose my

1:56strongest skills, available time, the

1:58work I like or dislike. So, when you go

2:00about this process yourself, you'll

2:01basically do a little bit of an

2:03interview with GPT to find out what

2:04you're actually good at. Then, you move

2:06on to stage two. Once you've gone

2:07through that process, GPT will recommend

2:10the specific business models that are

2:11suited towards you. And then stage three

2:13is a handoff where it will create a

2:15document that you can then use to spin

2:17off an individual idea. So the way I've

2:19set up this video is I have the

2:22identifying opportunity chat which is

2:23this chat where I'm doing an interview

2:25with GPT and then we're going to move on

2:27to the specific business model. So as

2:28soon as you find an opportunity you can

2:30also set up a project file like this.

2:32The good thing is now chatbt work allows

2:34you to access local files. So it's

2:35pretty much like having codecs now

2:36within the chat feature on your computer

2:39as long as you're using the latest

2:40version. And by the way, for this video,

2:41I'm using fast mode on GPT Astra High. I

2:44could move it up to ultra, but it is

2:46going to make things slightly slower.

2:47And we're not doing super complex work

2:50here. I do think when it comes to, you

2:52know, the real builds and, you know, if

2:53you want to flesh out like a deployable

2:55application, then you do want to use

2:57probably an even smarter or more capable

2:59version of GBT6. But for all the

3:01chatbased stuff, you know, even basic

3:02web design, planning, strategizing, like

3:05I think high is going to be good enough,

3:06at least from my experiments. And if

3:08you're willing to cop the extra token

3:10cost, then fast mode is very very good.

3:12Okay, so I basically asked me some

3:13questions. You know, I gave it some

3:14thoughts on what I was actually good at.

3:16So my skills, my personal preferences,

3:19how much time I have? So, you know,

3:20you'll basically answer all of this to

3:22GPT when you do this prompt yourself.

3:24And then it follows up with even more

3:25questions. So, what specific outcome do

3:27I think I could deliver? And if you

3:29don't know, you can literally just

3:30answer and say, "Look, I don't know. I

3:31need you to ask me more questions to

3:33work out what business model is suited

3:34to me." Because the reality is you're

3:36going to have certain skill sets that I

3:38don't have and I might have things that

3:39you don't have. So this is highly

3:41personal. For example, you could be

3:42great with people. You could be, you

3:44know, a great writer and that will shape

3:46the kind of skill set or moat that you

3:48lean into. You might not be good with

3:49people. You might be really introverted.

3:51So you might prefer, you know, more of a

3:53digital product model or more of a

3:54scalable model where you utilize UGC or

3:57other creators or, you know, creator

3:58marketing behind the scenes. Or you

3:59might be really hands-on. And that's the

4:01last example I'm going to give you,

4:02which is, you know, something that GBT

4:04can do really well. You might want to

4:05build a physical product and learn to

4:08develop, prototype one, and then source

4:09manufacturers to distribute one. So the

4:11whole point of the brainstorming process

4:13is to work out a side hustle that you

4:15can actually commit to. And to be

4:16honest, even if you don't make a ton of

4:19money, at least through trying a side

4:21hustle, at least through building

4:22something, at least through the

4:23experience of learning how to set up a

4:25website, a landing page, a deck, getting

4:27your first client, creating some content

4:29on it, you are going to become more

4:31skilled. And the irony is even if you

4:32don't make money off, let's say, a

4:34digital product, but you use AI to build

4:36a digital product, well, now you have

4:38the skill of building digital products

4:39with AI. So, and now you actually know

4:41how to prompt, and now you know the

4:42failures and why didn't work. And that

4:44is actually acrruable knowledge which

4:46can be then used for a consulting offer

4:48down the track. So the irony is the more

4:51failures you have of not making money

4:52with AI, the more skills you actually

4:54develop to make money with AI because

4:57consulting as a business model or or the

4:59agency business model is essentially

5:00just an arbitrage between what you know

5:03and what someone else knows. So you can

5:05basically close that gap. And the only

5:06way to develop the necessary learned

5:09knowledge is simply just to use AI every

5:11single day. So, I do highly suggest you

5:13use the prompt from today's video and

5:14you follow the steps that I'm going

5:15through and maybe one of these business

5:17models you actually like and you get

5:18inspired by cuz two of these I'm

5:20actually doing and I'll run through my

5:21approach to them as well today. And then

5:23after the interview process, it suggests

5:24to me a variety of business models. So,

5:27firstly, it thinks I should do a

5:29campaign pilot for established AI

5:31software companies because of my

5:32knowledge of marketing. Offer number

5:34two, it thinks I should create AI

5:36content workflow implementations, which

5:37I actually think is really smart because

5:38something that I've gotten quite good at

5:40is learning how to create content.

5:41Growing on social media, I have over 1.3

5:43million followers, I think, now across

5:45all platforms, you know, growing really

5:46fast on Instagram, YouTube, X. So, that

5:48is something I could probably package my

5:50knowledge into AI workflows and

5:51potentially sell it. Obviously, that is

5:52a, you know, unique skill set that I've

5:54acred. But that is the whole point of

5:56this process. AI is going to work out

5:57what your skill sets are. And don't

5:58worry if you don't have a skill set and

6:00you're like, I don't know what I'm good

6:01at. That's fine. and it'll work out some

6:02of the easier business models that you

6:04can get started with. And actually down

6:06below in my school community paired with

6:07this video along with the prompts, I'm

6:09actually going to give you a list of

6:10business models. I'm going to rank them

6:12by difficulty and I'm going to explain

6:14step by step how I would go about

6:16starting them. And these are business

6:17models that I have personally vetted.

6:18Either I've done them myself or I've

6:20tested them with others because, you

6:21know, I'm a believer that we should be

6:23using this stuff to better our lives.

6:24And obviously, one direct way we can

6:26better our lives is by creating side

6:28hustles, generating extra income, or

6:30improving existing businesses. And

6:31that's why I'm doing a lot more content

6:32on this now cuz it's genuinely how I've

6:34been using AI to optimize my business

6:36over the last couple of years. But

6:38especially now that these models are

6:40getting better, some of this just

6:41literally wasn't even possible 6 months

6:43ago. Like you're going to see,

6:44especially on number three, like the

6:46things GBT is able to do, especially

6:48with its visual capabilities, is it's

6:50honestly mind-blowing. So, all right,

6:51you've gotten to the end of the process.

6:52It's suggested a few business models.

6:54I'm going to go through three of them

6:56myself that I've handpicked. And once

6:58you've gone through this process, you

7:00know, there's probably a business model

7:01that you'll walk out with after the

7:02discovery phase. And the thing is, you

7:04might start going down the path of doing

7:05one of these business models and realize

7:07it's not for you. And if you already

7:08know a business model you want to do or

7:10you have a rough idea like, oh, you

7:12know, I want to do consulting. I want to

7:13create digital products. Like you can

7:14even skip this step and just go even

7:16more direct with GBT and just say, look,

7:19I think I want to create a digital

7:21product. I just don't know what it

7:22should be about. And in the example I'll

7:24give, which is example number two, I'll

7:26show you how I would actually go about

7:27building a digital product. And it's the

7:29same system that I'm using to scale my

7:30digital product, which has already done

7:31over $5.5 million in revenue, and we are

7:34experiencing a significant uptick now,

7:36and we'll continue to do so because of

7:38our AI implementation. All right, so

7:40business model number one, consulting.

7:42So, I have a starter prompt here, and

7:44once again, everything will be available

7:46in the free school community down below,

7:47which you should definitely join because

7:48there's the hub where I have everything

7:50from these videos and more to help you

7:52guys level up your lives with AI. But

7:54the prompt says, I want to start an AI

7:55consulting business, but I have no

7:57consulting experience, clients or

7:58context. Help me work out who to help, a

8:01simple service that I can learn to

8:03deliver, what schools and tools I need,

8:05how to find my first client, what to

8:06charge, and how to deliver the work.

8:08Then, a longer term playbook for

8:09actually scaling the offer. then ask me

8:12a few questions, I answer them. So, I

8:13give it my strengths. You know, I

8:15literally just voice prompt. I use

8:16Whisper Flow. I basically just speak

8:18into the microphone like this. The

8:20transcription picks up exactly what I'm

8:22saying. So, I don't even have to waste

8:23time typing. I can just have a

8:24conversation. Obviously, GPT also has

8:26voice. So, if you click on the bottom

8:28right hand corner, it's going to open up

8:30voice mode. And now I'm actually, you

8:31know, speaking to a voice agent where

8:33I'm able to have a conversation and and

8:36I don't even need to waste time typing

8:38or even reading text. So, you could

8:39literally go on a hike and just be

8:41talking to your phone about business

8:42models, going back and forth and, you

8:44know, honing in on things. But

8:46obviously, as you'll see, if you want to

8:47actually create, you know, a good deck,

8:48a good presentation, a good landing

8:50page, and a good offer, offer number two

8:52is actually, you know, a digital product

8:53offer, you're going to need to see it,

8:54tweak it, play around with it, and

8:55there's nothing better than sitting down

8:56in the office like this, you know, for

8:58executing something like that. Um, so

8:59after answering the questions on the

9:01consulting model, it gave me a

9:02recommendation. It thinks I should do an

9:03AI assisted inquiry and follow-up

9:05service for small businesses. Now, it

9:07specifically mentions small residential

9:10cleaning businesses. I think this is

9:12actually a very good niche. Like a lot

9:13of these local businesses, so let's say,

9:15you know, a cleaning business, a goods

9:17business, a lawnmowing business, a

9:20landscaping business, like all of these

9:21and pretty much any local business or

9:23any business in the world basically has

9:25the same constraint. Like unless they

9:27have so many clients and they can't

9:28fulfill, the majority of businesses have

9:30the opposite problem. They want more

9:31clients. And even businesses that are

9:33already at capacity, more clients will,

9:36you know, push them to hire more

9:37resourcing to increase their capacity.

9:39That's just basically the law of

9:40economics when it comes to business and

9:42entrepreneurship. So I do think starting

9:44with small businesses is a great way to

9:45go. But of course, it's good to stress

9:47test things. So although I agree with

9:48this model, I didn't want to limit

9:50myself too much to just this business

9:52type. So it gave me a lot of details.

9:54Actually, it pretty much gave me exactly

9:56on day one how to find buyers, how to

9:58understand the problem, how to build a

10:00demo, which is amazing. Like the concept

10:02I'm trying to teach here is that you can

10:03be a self-learner. You can go from not

10:05knowing anything about a business model

10:07to truly understanding a business model.

10:09You can even ask GPT for examples and

10:12you can even get it to create visuals to

10:14visualize every single step of the

10:15process. Now, it doesn't mean

10:17everything's going to be correct. You

10:18have to use your brain as well. You have

10:19to use your Nouse. You have to stress

10:21test things and where possible you have

10:23to speak to people and cross reference

10:24with other models. But it's going to be

10:25a damn good start. And even with my

10:27businesses, I run through my business

10:29models through GBT. And sometimes I'm

10:31shocked at what it's able to come up

10:33with and you know I find gaps in my

10:35business model or small tweaks that I

10:36can make to make more money. It

10:38literally happens all the time

10:39especially on number two which I'll show

10:40you the connections that these models

10:42now have to actual digital product

10:44providers and payment processes enables

10:47you to do a lot of the work within the

10:49terminal now which I think is really

10:50really cool. So I said look it's a good

10:51concept but you know cleaning is a good

10:53example but I want to go even bigger

10:54than that. I want to keep it broader

10:56because I think this same concept can

10:57apply to multiple local businesses. So,

10:59this is how I think the flow should go.

11:01I want to test it on family and friends

11:02first, build up some testimonials, use

11:04that as social proof, gain experience,

11:06because once I get local jobs, I can

11:08maybe document that process online and

11:10use that in advertising to help grow my

11:12following even further. And it stress

11:14tested my order of things. And then it

11:16sent me a finalized order of actually

11:18how to go about the process. But if

11:20you're like me, reading a slab of text

11:22isn't always very useful. So I actually

11:25went and asked it to create an Excal

11:27which is a mindm map tool so I can start

11:29building a real workflow and using that

11:32as basically an SOP and a business plan

11:34that I can edit over time. So there's

11:36actually an Excal connector. You can

11:38also ask it to do a mindmap style for

11:40you. It doesn't need to be on Excalor.

11:42It's just one way that you know you

11:43could own it and keep making

11:44modifications to it, but any visual mind

11:46map if it creates an HTML will

11:48theoretically work. So I'm getting it to

11:49recreate it now. You can see it pops up

11:51and this is essentially the business

11:53model. So you know in just a few minutes

11:55of going back and forth we have gone

11:57from nothing to a real business model on

11:59consulting. So the first step is

12:01obviously practicing something like this

12:03for example testing AI workflows testing

12:05using AI and building products building

12:07a fictional demo to test with family and

12:09friends. Like one thing you could even

12:10do is yeah go to a family member or go

12:12to a you know family or a friend that

12:14owns a local business and just ask a few

12:16questions about their business and try

12:17and just solve one problem from the

12:18business. For example, if they need new

12:20leads, look at how are they getting

12:22leads. Can you build a better AI powered

12:24CRM with HubSpot, for example, to help

12:26them get more leads through the door?

12:28Could you help them with their

12:29follow-ups? Are they, you know,

12:31following up on voice calls to their

12:32clients? Or could you create a voice

12:34agent for that? Really, consulting just

12:36comes down to understanding the

12:37business, understanding where the pain

12:38points are, and trying to use

12:40automations to solve those pain points

12:42and unlock more time and resourcing

12:44across the business. Once you've done a

12:45bit of practice and you have a feel for

12:47it, then you can move on to a real

12:48business pilot, which I'll show you in a

12:50second, the landing page and all the

12:51documentation for that we're going to

12:52create because doing a business pilot is

12:54going to help you earn genuine proof.

12:56And that proof you want to then put on

12:58your website as a testimonial. If you

13:00have a personal brand like me, then it

13:01becomes a bit easier because you can

13:03market online. But I'm cognizant of the

13:05fact, and this is the whole point of the

13:06video, that people are starting from

13:07scratch without personal brands. So, you

13:09want to get some real life social proof

13:11first and then you want to create

13:12content on that. I find the best way to

13:14create content is when you've actually

13:16done something. You know, a lot of

13:17people, they create content saying

13:18they've done stuff or run businesses.

13:20I'm a believer in the content is going

13:22to be a lot better and come across as

13:23more authentic if you're documenting a

13:25real story. Even if you did like a

13:26ladder challenge on Instagram, for

13:28example, where you document going from

13:30$0 in consulting revenue to 10 20k a

13:32month, people are going to join along on

13:34the journey. And if you do succeed,

13:35you're going to have all this historical

13:37social proof, which is going to really

13:38help your brand. And obviously once you

13:40start brand building that's going to

13:41help you scale up your business. And now

13:44instead of just relying on outbound

13:46leads and local clients and cold calling

13:48and cold emailing now you have you know

13:50warm inbound coming in because you start

13:52to generate demand that way. And that's

13:54what I found like I haven't even you

13:55know really put much effort into scaling

13:57my consulting offer. But on the small

13:59test we've started doing, we've got an

14:00inbound demand for this and now it's a

14:02decent revenue stream for me and it's

14:04one that, you know, I'm going to

14:05continue scaling up with my team because

14:06we're good at AI and there are people

14:08even, you know, tech forward businesses

14:10that simply want help with

14:11implementations or are willing to pay a

14:13bit extra for faster implementation

14:15because they don't necessarily want to

14:16focus resourcing on AI. You know, they

14:18would rather just focus on what they're

14:19good at. If you are a collectibles

14:21business, you know, focus on sourcing

14:23goods. Don't focus on AI and pay someone

14:25else to do it. And that's where you come

14:27in in this example. So I think this mind

14:28map is correct. If there's any part of

14:30it that you disagree with though, you

14:31know, you can edit this. This is

14:33ultimately your road map to starting a

14:35consulting business and you can tweak it

14:37however you want. So after we created

14:38the mind map, I asked it to create a

14:40launch kit. I asked it for all resources

14:42needed for me to get started from

14:44outreach copy to coming up with a

14:46business name to a landing page that's

14:47credible. Now obviously a lot of you

14:49will be able to think of a better name

14:50and a lot of you will be able to maybe

14:51give a reference to another website for

14:54a landing page, but I just wanted to see

14:55what GPT could do from scratch. Now

14:57remember, I did this in one prompt. So I

15:00obviously could build an amazing CRM and

15:02deploy it online if I spent time on it.

15:04This is just me going from start to

15:07something workable. But once you have

15:08the basics, then you know you could take

15:10a spreadsheet like this and make it as

15:11fancy as you want. But this is

15:12essentially a CRM where you list your

15:13businesses, service types, contacts,

15:15just a very basic CRM with a pipeline.

15:17And you could automate this on HubSpot

15:18once you become bigger. The next thing

15:20it created was the website. Now actually

15:22I didn't love the original website. Now

15:25I think it's okay. Like I think this is

15:26solid. All I ask is to make a website in

15:28a name and the pricing is okay. It's

15:30actually better than probably most

15:31websites online. But what I did and this

15:33is a really good tip for you guys is I

15:35went onto a website called godly.design.

15:38This is a website where there's actually

15:39a lot of free I mean some are paid but a

15:41lot of free websites which are like

15:43special. These are curated galleries by

15:45designers that are like you know really

15:48moving the needle in terms of AI design

15:49and website design in general. So, I've

15:51actually, you know, done launches and

15:53we've worked with clients where, you

15:54know, I've just referred them to Godly

15:56Design and we've used Godly Design for

15:58amazing websites from scratch. This is a

16:00hack actually because what you can do is

16:01you can give this to GPT or Claude,

16:03whatever you're using, but for this

16:04video example, it's GPT. I basically

16:06said, remake the website. I like this

16:08style. I just literally sent it a link

16:10to something on godly design. And then

16:12it redid just in one shot the entire

16:14design. And now my website looks way

16:16better. Get back to customers. Get on

16:18with the job. clear replies, useful

16:19templates, and a simple way to follow up

16:21on quotes set up around your business

16:22with you in control of every message.

16:24So, obviously, the exact consulting

16:26offer that you have will change things,

16:29but I think this landing page looks

16:30really good. You know, obviously you go

16:32back and forth like what I would add

16:33here are some more testimonials. I would

16:35add some animations to make it look

16:37sleeker, but and I got a template, a

16:39very, very good design that I've created

16:41in just a few minutes. You've already

16:42got pricing and yeah, we all know AI is

16:45very, very good at web design. I don't

16:47think that's a surprise to anyone, but

16:48I'm just showing you how quickly you can

16:50go from idea to iteration. And even just

16:52being hands-on with a website, I think

16:54adds a level of seriousness to a

16:56business model that you're vetting

16:57because it forces you to think when you

16:59have to create copy for a website, you

17:00actually have to understand the business

17:01model. So, it's kind of like chicken and

17:03egg. It actually forces you to refine

17:05your model because you're trying to sell

17:07a model to customers and simplify it to

17:08customers. And through simplification,

17:10you can find out a lot about a business

17:11model. It also created a launch kit, a

17:13client inquiry tracker. You can

17:14basically just keep going off this chat

17:16and building out your business and your

17:18operations. And because GPT also has

17:20computer use, it can actually open up

17:21its own browser, could even control your

17:23computer and do things on your behalf.

17:25So that is consulting. Hopefully I I'll

17:28do probably a dedicated video on

17:29consulting in the near future, but

17:31hopefully that helps you get started.

17:32Now, the second business model, this one

17:34is uh one that I'm focusing a lot on and

17:36it's probably the most scalable of this

17:38entire video. I'm actually doing both

17:40consulting and number two. I'm not doing

17:42number three, but I really wanted to

17:43show you guys what GBT could actually

17:45do. And I tried something more

17:46experimental, and it completely blew my

17:48mind. But number two is something I'm

17:49very familiar with, and this is digital

17:50products, where I've sold over $5

17:52million myself of digital products.

17:54We're scaling that hopefully to eight

17:55figures and beyond. If you actually

17:57include creators that have helped, maybe

17:58I'm already there, but you know, in

17:59terms of what I've actually made, it's a

18:01tad over I think like 5 a half million.

18:02What digital products are are

18:04essentially digital offerings. This

18:05could be software, so SAS. This could be

18:08tools, online tools, subscription

18:10services, info products. So, you know,

18:12packaging or curating information for

18:14people. This could also be agents

18:15technically. You know, you could package

18:16and bundle something I'm doing for my

18:18finance product. Agents and indicators

18:19and things that people that are

18:21interested in a certain market could

18:22use. So, you know, if you're in trading,

18:24it could be a trading indicator. If

18:26you're in social media, it could be a

18:28social media ideator skill or dashboard

18:30that you build. You can really go all in

18:32on one niche and and vet it out. But

18:34assuming you didn't know what niche to

18:35start in or what to do, you could just

18:37ask GBT. You could say, "Look, I want to

18:38create and sell a digital product, but I

18:40have no existing audience or customers.

18:42Could you please help me work out who to

18:44sell to, what a useful digital product

18:46is, how to test demand, and what to

18:48include, what to charge, and you know,

18:50ask me about my skills and my budget and

18:52my constraints so we can work out the

18:54best product for me." And one thing that

18:55I personally added, and of course, this

18:57will all be available in the school for

18:59you guys, so you can just copy and

19:00paste, is that I wanted to use

19:02which is my preferred. This is where I

19:04do all of my digital products

19:05development site and you know

19:06deployer/payment processor for this and

19:09I said I prefer the W CLI which is

19:11basically a way that you can manage your

19:13digital product from within a harness

19:15for example codeex which we're using or

19:17cloud code if you want to use cloud

19:18code. So that will actually help us

19:20build within chatbt the product the

19:23pricing and create the checkout. So

19:24firstly it asks me like what I'm good at

19:26what I'm interested in and I ultimately

19:28think if you're going to build a product

19:29you should be interested in it. One

19:30thing I'm interested in is trading. It's

19:32something I do in my spare time. Some of

19:34you guys might be interested in it. So,

19:35I said, "Look, I'm into trading." You

19:36don't need to be interested. I mean, you

19:38could just be savvy and have a business

19:40idea and know the target market. So, you

19:42know, you don't even need to be into

19:43trading to build a trading product, but

19:45it certainly helps or at least if you

19:46know people in that niche, it certainly

19:48helps. So, I said, I'm thinking of

19:49building maybe a tool for traders, some

19:50sort of indicator type thing, but I

19:52don't know exactly what. I basically,

19:53you know, gave it a few bits of

19:54information about me. It said its

19:56provisional recommendation is a focus

19:58trading view alert tool for people who

19:59trade around a day job. And I barely

20:01gave it any context as you can see. But

20:03I think this is actually a really sharp

20:04observation because that's the majority

20:05people. Most people trade around a day

20:07job. Most people want trading

20:09simplified. They basically just want to

20:10be told when to buy or sell. Now

20:12obviously if you wanted to go deep with

20:13this, there's back testing involved.

20:15There's obviously securities laws.

20:16There's lots of things that you would

20:17need to get right. But as a general

20:20principle, some sort of alert system,

20:22you know, would be a good product. It

20:23would at least attract a lot of people.

20:24So it gave me a few ideas. Breakout

20:26retest scanner. Weekly trading mistake

20:28review. I don't think this is as good.

20:29futures trade planning calculator. I

20:31think these two things are add-ons to

20:32number one, which is the winner. Says I

20:34should price around $29 a month, but you

20:36could do a bundle with other products as

20:37well. So, it gave me basically a layout

20:39of how to do it. But I said once again,

20:40look, I'm a visual learner. Like, give

20:42me a visual representation of this, like

20:44how I would actually go from, you know,

20:45not knowing about digital products at

20:47all to having a digital product. And it

20:49basically for this specific example

20:50outlined the eight steps from finding

20:52the problem to expanding the product and

20:54doubling down on what works which is you

20:56know something I always recommend in

20:57product or content iteration. So you

20:59know you have a full system here but all

21:01of this is theoretical. So I said make a

21:02sample indicator. I want to be able to

21:04put this into trading view myself. So it

21:06actually created a sample indicator.

21:08This is the pine script. And I'll

21:09actually show you something cool. I know

21:10not all of you are going to build

21:11trading indicators but I think it's

21:12quite cool. If you actually copy all of

21:14this and you select an asset, let's say

21:16gold, you click on the little triangle

21:18here and you click new tab and you click

21:23create new indicator. Then I can

21:26actually copy and paste what I just took

21:28from GPT. Click enter, publish the

21:31script. So I can click add to chart. You

21:33can see it's just loading the indicator.

21:35And now you can see it's actually

21:36created a break and a retest indicator

21:40which signals potential entries and

21:42exits at a certain zone. Now we don't

21:44know the back tested history of this. So

21:46if you wanted to get the entire history,

21:49you could go back up to GBT. You could

21:50ask it what it does. You could ask it to

21:52back test it. You could, you know,

21:54basically just go back and forth and

21:55create an indicator. I do separate kind

21:57of stuff like this on my finance channel

21:58if you want to check it out. I won't go

21:59too in the weeds in today's video, but

22:01the point is you can basically use GBT

22:03to create this stuff. We actually

22:04created a custom indicator for my

22:07community that I run called the money

22:09noodle. We have a new and improved

22:10version right now, but this is a trading

22:12system that works extremely well,

22:14especially in trending markets. And we

22:15developed this indicator and, you know,

22:17we've tweaked it to make it more

22:18accurate, better than most moving

22:19averages. And this actually, you know,

22:21provides a lot of signal for trading.

22:22You can see on, you know, Bitcoin, for

22:23example, it's propped the market up the

22:25entire way. And it acts as a bearish

22:27also rejection indicator. Now, we're

22:29actually getting a bit of a flip, which

22:30is interesting for the traders out

22:31there. But you can do a lot with AI now.

22:33And I think it opens up, you know, the

22:34possibilities for a lot of different

22:36products and iteration like this that

22:37you couldn't do before. So let's say I,

22:38you know, went back and forth. I created

22:40an indicator that I like for traders and

22:42I, you know, thought there was a value

22:43ad and I wanted to actually build this

22:44product out. I don't even need to really

22:46do much apart from signing up for a WP

22:49account, installing the CLI, and I just

22:50asked GPT to install the CLI. Then I

22:53asked it to teach me how to use the CLI

22:56to deploy a product. And you can see

22:58here, you can literally create a product

23:00within Now, I can't show the

23:02deployment now cuz I would need to

23:03create a new account and I'm going to

23:05probably do a future video on it. So,

23:06make sure you subscribe to the channel.

23:07It's going to be probably its own

23:09series. You know, I would need to create

23:10a new account, get it verified. So,

23:12obviously to actually deploy something,

23:14there is a little bit of work. But in

23:16terms of everything else from creating

23:18the checkout to the landing page to

23:21deploying the indicator to making

23:23adjustments to pricing to managing

23:24community, it can actually all be done

23:26within the CLI now. So essentially you

23:28can run a digital product directly

23:31inside GBT or whatever harness you have.

23:34So codeex cloud code whatever. Now what

23:36I also asked for is a website or a you

23:39know landing page that we could deploy

23:41on and I could you know just connect

23:43this with the CLI that traders would

23:45basically you know go to when we

23:46advertise. We could advertise through

23:48paid ads. We could advertise through

23:49creator marketing. We could advertise on

23:50the W marketplace. You can even manage

23:52ads on GPT itself which is very cool.

23:55And I gave it no guideline references. I

23:58didn't iterate on this at all. This is

23:59like the first time I'm really looking

24:00through it. And this is pretty damn

24:03good. This is this is very cool. This is

24:05a nice visual. Actually makes me want

24:07the indicator. It has a, you know, it

24:09get it gets you to the checkout very

24:10quickly. It's not making you scroll

24:12through a bunch, you know, to get to the

24:13get to the payment. It goes through the

24:15tool what it does. And, you know, once

24:17again, I did this in pretty much one

24:18shot. I didn't use godly design. I

24:19didn't use anything else. So, you can

24:21literally spin up a landing page and

24:22then you can connect it to and

24:23actually deploy it. really really

24:25quickly. And once again, I'll put the

24:26guide for all of these in the school

24:27community below. And then the third and

24:29final idea, which is the most ambitious,

24:31and I actually got this idea from

24:32Mickey, who I maybe by the time you're

24:34watching this, the podcast is out on the

24:35channel, but it should be very soon.

24:36Anyway, very good interview, very smart

24:38guy. He gave me this idea today, and

24:40then I went to test it out of the fact

24:42that GBT6 Astra just has such good

24:44visual capabilities. It's actually quite

24:46good for like 3D image modeling on

24:48Blender. And because this makes it a

24:49good model for engineering, it pretty

24:51much now enables us to engineer our own

24:53products. So the opportunities here are

24:55endless. Any physical product that you

24:57want to make, anything custom that you

24:59want to make, you can now create. So in

25:01this interview/podcast that I did this

25:03morning with Mickey, he basically said,

25:04"Look, I bought an Apple HomePod, but I

25:06realized that the HomePod, you know, I

25:07couldn't jailbreak it. It wasn't

25:08customizable. So I couldn't, you know,

25:10connect Siri to my agent. I couldn't

25:13connect it to Hermes or I couldn't

25:15connect it to Codeex, which has my

25:16personal, you know, brain. So it

25:18couldn't actually do anything. So if you

25:19want a Jarvis sitting next to you on the

25:20desk and you know you could speak to it

25:2224/7. It'll note things but it will also

25:25actually do stuff unlike some of these

25:26other tools that are starting to come

25:28out like these pocket devices that

25:29you'll probably see not the brown pocket

25:31but just like you know pocket size

25:32devices you're seeing. What they don't

25:34do is they they don't connect to the

25:35rest of your stack or at least you need

25:37to build like web hooks for all of it

25:38and it's not like fully customizable.

25:40Whereas theoretically you could build

25:41your own customizable stack front end

25:43and back end. Previously maybe all you

25:44knew how to do was backend. Now you can

25:46actually add a DIY front end. But

25:48obviously, you know, you could use this

25:49for any concept. If you were doing, I

25:50don't know, motor parts or if you were,

25:52you know, doing repairs or if you wanted

25:54to build literally anything. If I wanted

25:55to design a water bottle, a 3D printed

25:57water bottle, headphones, like you could

25:59probably DIY it all. But I think this

26:01was a good example to use for today's

26:03video because this is the type of thing

26:04that could be added as a physical

26:06offering on top of a consulting product.

26:07So I basically said, "Help me develop

26:09the business model." But I had this idea

26:11and it basically went through the entire

26:12business model and I asked it create

26:14Blender sketches and sourced the parts

26:16for me and it came up with an entire

26:19website, an entire design which I'll

26:21show you around a personal assistant

26:23called still that sits in the room. It's

26:25a voice assistant. It's thoughtfully set

26:27up around you. It has routines. You can

26:30teach it. It, you know, manages your

26:31devices. And, you know, this is the cost

26:33for a custom setup. So, you could, I

26:34don't know, in theory order the parts

26:37and ship it to a potential customer.

26:39This is a custom hardware which I think

26:40is very cool. So that's the landing

26:42page. But what is very cool is it can

26:43actually create sketches on Blender. I

26:45don't have Blender downloaded but I

26:46literally have the files where it gives

26:47the product spec. You could get this

26:49hardware custom manufactured. What I

26:51actually did or you could use, you know,

26:52a 3D printer and I'm sure you could get

26:54it to do the sketches for that. What I

26:55also did is I asked it to source

26:57suppliers and it went through and it

26:58found the cheapest suppliers for all of

27:00the individual parts that were needed.

27:01So, you know, what powers it, the power

27:03supply, the chips, the hardware,

27:06manufacturers, the, you know, the

27:07integration with Wi-Fi, you know, voice,

27:09microphone, etc. And, you know, I could

27:11ask it something like, "Hey, GPT, could

27:13you please model the 3D sketch to show

27:17each individual component and how it

27:19connects to each other in terms of the

27:20internal hardware, you know, combining

27:22with the external shell. So, okay, the

27:24prompt's done. And you can see here it's

27:26mapped out the full design with the

27:30labeled components, which is super cool.

27:31I saw someone do this for like cars and

27:33watches the other day, which is super

27:35cool. You can pretty much break down

27:36anything, you know, if you're just

27:37curious in how things work. But yeah, it

27:39breaks down every single part. It also

27:40breaks down the hardware components and

27:42it has every single component labeled so

27:44you can essentially build your own

27:46hardware. Now, I'm not saying this is

27:48easy. I'm not saying, you know, anyone

27:49could do this build. I mean, you

27:50probably could if you went back and

27:51forth with GBT, but you know, obviously

27:53there's, you know, but obviously there's

27:54soldering metal and, you know, skills

27:56you would need. But the point is that

27:58things that previously didn't seem

27:59possible are now possible with AI and we

28:02could just get a much deeper

28:03understanding of how the world works and

28:04how to build things from scratch. So,

28:06essentially in this chat in just a

28:07couple of minutes, we've gone from an

28:08idea to a business model and the

28:10physical hardware design and the part

28:12sourcing to create a physical product,

28:14which I think is pretty damn cool. So to

28:16tie things together here now with AI,

28:18you can diagnose the best business model

28:20for you. You can choose the best one.

28:22You can validate and test demand. You

28:24can build and offer. You can build and

28:26offer a product with AI. And then you

28:28know the step that I didn't show you cuz

28:30this would be a 3-hour video if I did is

28:32the fact and we cover a lot of this on

28:33the channel. Now you can build the

28:34workflows once the business is set up to

28:36automate processes within the business

28:38and you know keep iterating and

28:40reviewing data so you can keep

28:41improving. And I'll leave the business

28:43models that we spoke about today, plus a

28:45few more cuz there are probably 10 that

28:47I think are viable in the AI era with

28:49exactly how you can get started with

28:51them alongside the prompts for you to do

28:52your own discovery process in the

28:55description below. It'll be under the

28:56school community. If you go into the

28:57school community and you click on

28:59classroom and then go free assets

29:01library, we have the full library. Just

29:03go to the corresponding video and you'll

29:05be able to claim the skill. or if you

29:06just go to the community tab, we pin

29:08every new video and if we have assets

29:10for that particular video, they'll

29:12essentially be linked here. So that is

29:13where you can find it. Come be a part of

29:14the community. Thank you for taking the

29:15time to watch this video today. Your

29:17homework is to go and at least get a new

29:20side hustle off the ground. And I'll see

29:22you next time on the channel. Peace out.