let us start our study session for today. The first hour, I'm going to be studying some business stuff. In the second hour, I'll be working on some AI projects, and we'll be using GPT-3 to see if it speeds the studying up and makes me more productive. By the way, this is Scar out [Music] Morocco, and I'll show you guys around later, but first let's get some studying done. [Music] Okay, by the way, at the time of filming, most people don't have access to GPT-3 like the voice functionality, and I don't have it either. Maybe by the time you see this video, that would be there, but I wanted to try it out early. So, I teamed up with Posit to build a shiny app in order to get the voice functionality. It's not perfect, but it's pretty good. Hey, I'm ready to get started studying, but I'm really not feeling like it. Could you give me some words of motivation? There is a 20 to 30-second lag here for this app, which allegedly

the real GPD 4 would not have FYI. I will be editing it out for the rest of the video though. Hey, I know it's tough getting started sometimes, but you've got this. Think about how awesome it's going to feel once you've accomplished your study goals. You are capable and strong, and every step you take gets you closer to where you want to be. Believe in yourself and remember it's all about progress, not perfection. You've got this, one small step at a time. Keep going more a fantasy so w you stay here only. [Music] While okay, so I want to start a new membership-based monthly program in which people can learn AI skills, and then they can also work on building AI products. I listed out these steps in terms of what I think I need to do in order to create a business plan. First of all, can you read these out to me? Sure thing. Here are your steps: 1. Research different resources on how to start membership. 2. Write notes on pros and cons of each membership platform. 3.

figure out some important topics related to AI skills. Four, find some good examples. Five, create a plan. Six, determine a Time Li. You've got a solid out. It looks like you're really organized and ready to get things rolling. Best of luck with starting your new membership program. Okay, so for number three, you got it wrong. I actually wrote figure out some components useful SL. Checkout guides. Um, anyways, so what else do you think I should consider? Hey, I noticed I got number three wrong from your list. You wrote figure out some components useful SL. Checkout guides. Got it. For additional steps, you might want to consider: one, creating a marketing strategy to attract and retain members. Two, setting up a payment system for subscription fees. Three, developing a feedback mechanism to continuously improve the program. Four, planning regular content updates to keep members engaged. Five, establishing a support system to handle any issues members might face. These should help you cover more ground as you set up your membership program. Actually, could you help me come up with a business plan for this specific product? Creating a business plan for

your AI skills membership sounds like a fantastic idea. Here's a quick guide to get you started. One, executive summary. Start with a brief description of your business idea. Explain your goals and what your membership offers, like the AI skills projects and new technologies. Also, mention your target market and what makes your offer unique. Okay, so before I continue into the video, I first want to introduce you guys to the sponsor of this portion of the video, magical. Thank you, magical. Magical allows users to connect any app or site above and beyond what Zapier or other IP providers can connect. It works by allowing users to label any information on any app and move that information to another app or tab. For example, you could label all the restaurants on TripAdvisor and move them into a sheet. You can choose what in the page you're interested in, such as the name, the address, and the cuisine. And then after that, here's the magic. Continue to automation, add to a new spreadsheet, click all of the tabs, and voila. And this is just one use case with the free magical AI browser.

extension. We can now finally automate all sorts of repetitive, brain-numbing tasks like automatically populating forms and spreadsheets, responding to emails 10 times faster using personalized templates, and drafting contextual emails with just a single click. This saves me at least 7 to 10 hours per week. Magical works on every site, including internal company tools, and connects to any app with no code or setup. Sign up for free to save time on 10 million plus websites by going to this link. Over here also linked in description. Thank you so much, Magical. Now back to the video. Could you repeat um, financial projections? Actually, can you just speak slower? Here's a slow breakdown of the financial projections section. I just want to point out that this is one of the limitations of our app. The real GPD 40 should be able to vary its tone a lot better, like actually speak slower. What are some different platforms, uh, people usually use to host memberships? All right, there are some common platforms people use to host memberships. One, Patreon, great for creators to offer exclusive content to their supporters.

two thinking ideal for educators to create and sell courses three teachable another I want to have payment that's already embedded in. I would also want there to be a way in order to live stream and to put Links of project. Oh yeah I should have mentioned that my main line of business right now is social media through YouTube. If your main line of business is YouTube, then YouTube memberships could be a really smart choice. It's built right into your platform so your audience doesn't have to go anywhere else to sign up and access exclusive content, plus it's super easy to set up and manage. What are some potential cons of using YouTube memberships? First off, YouTube takes a 30% cut of your membership Revenue, which is quite high compared to other platforms. Plus, you're limited in terms of how much is it for other platforms that you mentioned think if it has a free plan with basic features and their paid plans start at about $39 per month. For the $39 per month, is it based upon um Revenue that we're making or is it

just a flat out $39 per month. Okay, so first impressions. Overall, having that voice, The Voice, um, interaction is so much nicer than just having that text. Um, I'm actually really surprised. I thought it would be kind of more tedious, but it really is so much nicer working through it. Also, being able to ask questions, um, as we're going through it, asking it to repeat things and things like that, it's just a lot more of a natural way in order to do the workflow, as opposed to communicating with CHPD through text. Um, and then the visual feature is also really nice, being able to showcase it and it can read things relatively well. I can see how if there's like certain diagrams or certain things, um, as I'm going through the plan, I might just have it on my phone, a certain diagrams, and I would just ask it directly to explain it. Um, yeah, I'm sure all of that will come into play later on. Not in terms of limitations, there are limitations just because we built this using a shiny app as opposed to the

actual GPD 40. So if you are going to be using it by the time this video comes out, I think this will be really nice. So first of all, is going to be you can't see the screen. So when I'm typing stuff, I'm not able to show the screen. Um, to GPD 40. And they did show you can do that on a demo though. In the code, there's this function Foo. Can you describe to me what the plot would look like if I had versus didn't have this function? The function Fu XY takes in a data frame X containing temperature data. Also, it doesn't have access to search the internet, which again, once GPD 4 is out, it should have access to those too. Uh, another limitation is that the tonality of it. Um, the voice is really natural for sure, but the tonality that they show in the demos, that's not able to be replicated through this. Um, app as well. Hey chat, GPT. I'm Mark. How are you? Oh, Mark, I'm doing great. Thanks for

asking, how about you? And you know, I can go into more technical detail, but you can also just check out the GitHub for exactly what it is. It's doing very briefly. It's going through Whisper for the audio and then GBD4 for the video aspect. That's why it doesn't have. That's why it's still translating the voice to text. And that's why you don't have like tonality and things like that. So this should all be much, much better with the actual GBD F as opposed to just going through what we have right now through the API. Anyways, I definitely enjoy working with CBT through this method as opposed to purely through text though. It's really nice. All right. So next step is I'm going to figure out how to build some AI projects. So using a platform called Kestra. I wanted to like figure out how their platform works, try out some of the flows and things like that. So I'm going to see how good it's going to be to help me.

figure out how to get started and start building things and fix any bugs that I could potentially have. Okay, before we start our next study session, I need to take a break. I'm getting tired, so what I'm going to do is, since I'm in Morocco right now, um, I should learn either Arabic or French. I'm going to go with French because that seems more closely related to English and probably easier to learn. So I'm going to test out GPD 4o in its ability to help me, um, learn different objects. Hey, I'm trying to learn French right now. I'm going to show you a series of things, and can you tell me how to say them in French? What is on screen right now? On screen right now Isen [Music] [Music] that's actually very good. There is a field, there is an apartment construction, the sky is blue, um, it's a sunny day, and we're out, uh, exploring the nature, something like that. It's actually very good. How do you say the objects on

screen in French holding a camera on a tripod standing on a dirt path surrounded by dver grass. In French, you would refer to these objects as the person, the camera (CA), the tripod, the dirt path, the shim, the dry grasses, the buildings in the background. Letim, how do you say the object on my feet in French? In French, which means boots, so you can say how would you read it, uh, I'm probably like six seven or pronunciation wise, it's not like, can you say it yourself? I mean the both like something that and it's saying like B. I don't know how do you say the object in the video in French? Object you're likely referring to is UN tunnel or unpass suine. French, what is this in French? In the video, there's a train passing by in the background. In French, you would say un time. The sun was getting really hot, so I decided that I learned enough French and it's time to go make some AI projects. So first things first, what do

you know about kestra. From what I know, kestra is an open-source orchestration scheduling platform designed for running complex data workflows. It's built to handle tasks involving data processing, integration, and ETL (extract, transform, load) operations. You can define workflows using YAML files, which makes it quite versatile for developers. How is kestra different than, say, scheduling cron jobs? Alright, so kestra is different from usual job scheduling systems because it's more about managing data workflows, while traditional job schedulers might handle, say, running a script every night at midnight. Kestra steps it up by dealing with complex data pipelines. It's like the difference between a simple alarm clock and a full-blown project management tool for your data process. Got it. Can you remind me again how YAML files work and how does it work in relation to kestra? Alright, so YAML files are basically a way to format data that's easy for both humans and machines to read. YAML stands for YAML Ain't Markup Language. It's used to write configuration files in a plain text format with a structure that's

hierarchical in relation to kestra. Then I had some back-and-forth conversation about yam files and how they relate to kestra, which was super helpful, and I was able to understand things a lot faster than I normally would have if I was just looking through documentation myself. Does that make sense? Yep, that makes sense. Can you give me some concrete examples of how it can be used? You mentioned something about data workflows and streamlining those. So what are some concrete examples where this would be useful? All right, so here's how kestra could be concretely useful. Examples were also super useful in getting a deeper understanding and brainstorming what kind of projects I can build. Ch40 did help me figure out what was going on, and I started installing Docker, but I also did start losing my mind a little bit because it was taking forever to install, probably because of the slow internet. There are only 4 minutes left, and I've been installing Docker for 60 minutes. That's literally what I did for my second study session. Try to

install doer. I don't understand why Docker won't load. You're literally making me lose my mind right now. Okay, I need more motivation from GPT-3. Hey, I'm feeling very stressed right now because I can't get Docker to install, and I want to cry. Can you give me some words of affirmation? Sometimes things just don't go as planned, and that's completely okay. Remember, every problem has a solution, and you're capable of finding it. Take a deep breath. You've got this. Keep believing in yourself, and you'll get through it. Remember, it's okay to ask for help or take. Oh my god, it worked. Localhost won't load. I got it. Alright, okay. So here's the goal. I don't to do this. Our time is technically already up, but I insist on at least getting something running for this study session. I'm probably going to have to do. Remember, when things get tough, your inner strength. Go to sleep. Okay, so yeah, this is not working out. So I need to like test out these AI projects.

like build some of these AI projects later, but at the very minimum to make myself feel better, I just like need something to work, you know? It's just like something must work, even if I just manage to open it and type like one line of code, I would feel better. Okay, well, since it's, I don't know if it's going to download, but you know what? I'm just going to try to get through some of the code right now and see if we can figure out how it is that I can run it. After I managed to get this thing to download, I'm going to show you a yaml file and can you describe to me what's happening in it. So, in the video, you're showing a phone with a yaml file displayed. In this particular yaml file, it seems to set up some tasks related to data warehouse and analytics with various steps including clone. Can you describe it to me line by line? All right, in the video, I see you're showing a yaml file on your phone.

I'll break down what's happening line by line. Here it goes. One ID CSV files analytics. This seems to be the identifier for the task or workflow. To namespace tutorial. This specifies the namespace, which is kind of like a grouping or domain the task belongs to. We went through the YAML file line by line, and it was honestly so helpful. I got a really good idea from the detailed walkthrough. Okay, summary for coding section. Unfortunately, I wasn't able to get through most of it. It was helpful in helping me diagnose what the issue is. When I couldn't get a doctor to install. In the demo, it showed that it would be able to see your screen. So it wasn't able to do that. But I think if it did, it would have been very helpful in terms of being able to go through the code with you. I would be very interested in testing out how good it is to verbally write code. And I guess I'll get to that next time. All right, thank you so much for watching. I hope this was helpful. If

at all. And if you want to try out this application, assuming that GPT-4.0 has not come out yet, um, by the time of this video, you can do so. I'll put the link in the description over there. This app can only do maybe like even 30 or 40% of what the demos showed. I honestly think it's surprisingly useful just to have that voice communication and be able to have that video feed. I'm really excited to try out the actual GPT-4 official app when it becomes available to us. All right, I will see you guys in the next video or live stream.

Study With Me Using GPT-4o

Hi! My name is Tina and I'm a ex-Meta data scientist. Now I create content and other internet things! 

This channel is about coding, tech, career, and self-study. I love learning new things and my goal is to always minimize effort and maximize outcome! 

I strive to do things in the most efficient way possible and with the highest reward. I basically refuse to do extra work to achieve my goals c: 


