I signed up for Y Combinator's Startup School

Daniel Bourke finished a 100-page machine learning book, built his first iPhone app with Core ML, and launched his dog-breed classifier on A Cloud Guru for his first paid project. He's now enrolled in Y Combinator's Startup School to develop Nutrify, a nutrition-tech startup.
Chapters
Morning! It's cold this morning, so I got 3 layers on. 1, 2, 3. I got a meeting this morning. How beautiful is the internet, right? You can meet someone who's across the world almost instantaneously. Absolutely amazing. Could turn out to be a potential partnership, but we'll see how it goes. Yeah, beautiful. Well, have a good afternoon. Or evening. See you later. Oh my goodness. That is the power of the internet. You never know who you're going to meet. So the takeaway there is create things, share them online. You'll find people who are into it. Meeting went well. We'll see how it goes. All right. So you remember how a couple of days ago I finished the 100-page machine learning book? Well, I put together a little review of some of the main things that you should know about the book, like who should read it, should you buy it, etc., etc. Long story short, you should. So that's what I'm going to do now. I'm going to put together the video, edit it, get it up, and then do some study later this afternoon. 3, 2, no, no, that's good, but that's, that's the last part I need. Shooting for the 100-page machine learning book review went amazing. So now I've got to edit that out. Look at that dude, he's pretty cool, eh? So I'll edit that, upload it tonight, should definitely be out by the time you watch this video. And someone left a comment on one of my last videos about was I going to watch the Neuralink livestream? No, I didn't watch the livestream, but I did find a video summarizing Elon Musk's presentation. And it's literally blown my mind, pun the pun. If you're not sure what Neuralink is, it's basically one of Elon Musk's companies to inject AI into the brain. So I nerd out on that naturally. You should definitely check it out if you haven't. But after I finish this video, I'm gonna read Apple's Swift book. So they have a 1,000-page book on Swift and how to build apps, which is what I wanna do. I wanna use Swift for deep learning and Swift to build products. So I'm going to basically read that book end to end and see where it gets me. Hey Siri, 25-minute timer. Okay, your timer is set for 25 minutes. I'm going through the Swift book, or Apple's Swift book on iBooks. It's basically, well, it is for free, right? You can download it and it goes through the latest version of Swift, latest version iOS, latest version of Xcode. And it's written in such a beautiful and simple way, right? It's kind of inspiring to me to— if I was to write my own articles or own documentation on doing things, I want it to be in this way. So it's in language, it's like, here's a concept, here's an example. And so following on from what I said in the last video, I said, I'm not going straight to the target. I said, I want to build products and make applications. That's what I'm trying to do, what I'm diving into doing. And too often in the past, I've kind of like zigzagged around trying to learn something instead of like just, just getting a book on it and reading it start to finish and then implementing the knowledge. That's the thing with learning, right? Always have a bias towards doing, but don't forget, like, it's okay to take a moment, be patient, learn the foundations, then try to put them into practice. If you don't get it, read it again. If you don't get it again, read it again, try it again. That's the thing with programming. I think Swift had a— had a— this book had a great statement at the start. I'll put it up. Somewhere here and pause the video for a second so you can read it. And that's what it's about with programming. It's going to be hard. You'll get stuck, but then you hit a roadblock, you'll read something once, it won't make sense. You'll come up to a problem, you won't be able to fix it. You'll, you'll give it a few hours, you'll give it a few days, you'll come back to it. All of a sudden, whatever you read a few days ago has now kicked in and you can put it into practice. So I'm going to keep reading this for a bit and then I've got something. Hussam, Hussam sent me something really cool. So, I think I'm going to sign up, you know, just commit. I'll show you that in a second. Hey Siri, 25-minute timer. Your timer is set for 25 minutes. Another Pomodoro of reading first. Gonna build things, gonna learn things, gonna build things. Fridays are for learning, Saturdays are for building. Gonna make things, gonna lose things. Fridays are for building, Saturdays are for learning. I like to code, everybody likes to code. Oh, look at that, man! I've got an app running. It's got nothing on it. We've got our first app. Haha. Now I need to get my microphone for the computer, which is somewhere. Haha. First blank app. I caught my microphone, leh. Plug her in. Alright. Ladies and gentlemen, welcome to the show. It is your boy DJ Daniel. Let me take you to a land of sweet, sweet, sweet euphoria. A place where you forget your name, but I won't, because I know your name. Okay, we've run our first app. Tick. Now we just gotta— now we're just gonna build something great. Sounds so simple, right? All right, back to the boring part. Well, not boring part, it's fun. I get to read. After simulator has launched, you should see a device image with a white background. Boom! We've done it, we've done it. Ticking, ticking boxes. Look at that. You watch this space. Watch this space, baby. All right, would you look at that? First app run on my iPhone. Look at— look, blank white apps in progress there, right? That's top of the App Store right there. This could be the Keep Calm app, right? You open it and nothing happens, so you just chill out and don't use your phone. First app on the iPhone. That's what we're talking about. Core ML, baby. Integrate machine learning models into your app. Really liking this, this book and this, uh, this documentation. I mean, it's one thing to— that's what I missed out on, I think, learning other languages and programming in general is just not— just been too afraid to just dive into the documentation and read the, read the actual docs. And even if I didn't understand it, just read it and then come back to it. Read it, come back to it. I think I said that before. Okay, so we've done about an hour of hands-on Swift study, and people asked in a recent video, what am I using? Like, what tools am I using for my business? One of the main ones is Harvest. It's kind of how I'm keeping track of how I'm spending time. Like yesterday, YouTube video editing, today, learning Swift code. Now, again, this isn't perfect, right? This week, I haven't been good at tracking my time. Last week was a bit better, right? So, we got here, learning creating, reading, writing. Etc., etc. So I'm getting slowly better at it, but Harvest is just a simple time tracker so that I can go back and go, okay, here's what I want to do, how did I actually spend my time? I'm gonna have a little break, come back, do another about hour or so, and then, uh, yeah, and then we'll see where we go from there. Hooray! Just spent 10 minutes Lying down on the grass in the backyard doing nothing. Probably the most productive 10 minutes I've spent all day. Brain freed it up, looked at the sky, thought of the birds, thought how grateful it is to be here, to be able to learn, to be able to create. Let's get into a bit more Swift, hey? Then we're going to do a YC update. YC? What's that? Stay tuned. Alrighty then. That'll be enough Swift study for today. We made it up to structures, which I believe are similar to, similar to Python classes. That's enough studying for today, but massive news. Well, this is part one of the two massive news to wrap this video up, to wrap this week up, this amazing week that it has been. Check this out. Hi Daniel. I just wanted to let you know that your project has now gone live on the A Cloud Guru site. Congratulations! The email below is what I sent out to everyone. Please feel free to send through your invoice whenever it's ready. That's mean dollar dollar to you! Oh wait, it's a bit dark in here. Get some light. What that means is big dogs gotta eat. Speaking of big dogs, this is what my project was on. How cute is that doggie in the window? So essentially I made a video using Google's AutoML to classify different dog breeds, right? This is where I think machine learning is really going, right? It's merging with software engineering. So in the future, I reckon in 1 to 2 years, AutoML services will basically be available for practically any main machine learning task. Maybe even earlier than 1 to 2 years, maybe in the next year or so. You already got AutoML, AutoML Vision, AutoML Tables, AutoML Language. Etc., etc. So if you're going to get around something, get around technology, get around the capabilities of AutoML services. But the next thing is the final, final little clip of this video is Startup School 2019. I'm signing up. So my boy Hussam, thank you Hussam, sent me a link saying, Daniel, you should sign up to Startup School. It's this really cool thing where you can go through a free curriculum, learn how to build a startup, learn how to make something and get it live to the internet or get it live to the world, which is kind of the internet because that's how the things that I like to build get live. Anyway, what is it? It's a 10-week online course with Y Combinator. Now if you haven't heard of Y Combinator, that's where Dropbox and like Airbnb came from, was Y Combinator. So it's pretty legit. So basically what the 10-week course is, is you get a startup network, a forum, mentorship, exclusive deals. So you get some, some cost benefits with like Google Cloud, AWS and whatnot. And you have a chance at the end, if your startup is good enough, you can get $15,000 potentially in cold harsh cash. But I'm not really after that. The main thing that I'm doing it for I'm signing up to go through the 10-week process. It starts next week, right? It's to kind of kick me in the guts or kick me in the butt, a little bit of both, into building my startup, right? Which is, it's going to be Nutrify. I want to get into nutrition education, use technology, how to do that. But I'm not exactly sure how yet. So I'm kind of going to use this 10-week process as a way of figuring it out. And building it at the same time and learning it to build it at the same time. But that's the massive news. Signing up to Startup School. I'll leave a link below if you want to join. Starts next week, Monday, July 22. Whoo! Anyway, Bordy's got a message for you for the finish of this video. What's that? Where do we go? Keep learning, keep creating. Oh yeah! That's what it's all about. I'll see you next week. Hey, Seven, guess what? Guess what? We get to send through our first invoice as a business. Our first invoice. He's too distracted. Too distracted. Down! Down! They don't care. That's the simple life, right? Just play and enjoy. I just want to live more like a dog.
More from Daniel

Small Language Models (SLMs) Are the Future: Fine-Tuning AI That Runs on Your iPhone

Introducing Sunny - A MedGemma-Powered Skin Health Tracking App (Kaggle Competition Entry)

Benchmarking: NVIDIA DGX Spark & NVIDIA RTX 4090 | Part 4 - Comparing the results

Benchmarking: NVIDIA DGX Spark & NVIDIA RTX 4090 | Part 3 - Running LLMs on the RTX 4090

Benchmarking: NVIDIA DGX Spark & NVIDIA RTX 4090 | Part 2 - Formalizing the tests

Benchmarking: NVIDIA DGX Spark & NVIDIA RTX 4090 | Part 1 - Creating a series of tests

Local Multimodal RAG on the NVIDIA DGX Spark | Part 4 - Publishing our demo to Hugging Face

Local Multimodal RAG on the NVIDIA DGX Spark | Part 3 - Putting it all together


