i left my machine learning job

Daniel Bourke quit his machine learning job at Max Kelsen to make online video content and build a health education platform. He saved enough money and cut expenses to fund the leap. He's focusing on health, technology, and art content while figuring out what comes next.
Chapters
Very, very, very big day on tour today. I'll tell you why in a second. I gotta catch a train. You see, about 14 months ago, I was in this exact kind of position, and I live for the moment when things come full circle, you know. And what's the significance of this little place here? Well, I was walking down this street, and it's still windy as ever. It was my first day at Max Kelsen, starting an internship as a machine learning engineer. On the way to the internship, a little bit scared, but you know the same feelings that scared cause, or the same emotions that, the same hormones that cause you to be scared are the same ones that cause you to be excited. So that's what I'm gonna do. I'm gonna flip that scaredness on its head and start to get excited and go in and learn some things. Just caught the train here with my little brother. So thanks for the chat, bro. And yeah, I'll catch up with you guys either after the day or through the day if I can share what I've learned. And that led into a job, and now coming full circle, Today is my last day. We on? Yep, we're on. Okay, as you can probably tell from the start of the video, I left my job as a machine learning engineer at Max Kelsen, and the TL;DR version is I decided to try and do my own thing for a little bit. All right, well, that's enough. No, I'm kidding. I'll give you a bit more of a backstory. Alright, so long story short, I had a little bit of a gut feeling that was kind of telling me, you know what, Daniel, you should try and take a risk. See if you can go out on your own and start your own business. So that is essentially what I'm going to be doing. For— I was going to insert a timeline there, but I kind of don't have a timeline in mind. Let's say forever, because that is the most ideal case. And instead of thinking about what's going to go wrong, I've decided to start reminding myself what could go right. Because when I was writing my letter of resignation on the train, typing it out on my iPhone, in my head it was going, you know what, this job is really good. These people are really good. The problems you're working on are really good. You're learning lots of things. Why the hell are you typing a letter of resignation? So it was arguing with my gut and I can't really decipher what my gut was saying, but I had this little feeling in there that's just like boiling up and it's just doing these circles and it's kind of like, are you going to listen to me or not? It's kind of like someone was constantly, imagine someone constantly walking around poking you going, hey, Hey, you doing that thing? You doing that thing you said you were going to do? You know, that thing you really want to do? And that's kind of what I'm doing now, doing that thing I really want to try and do. Well, I really want to do. I've got to stop using the word try. I should have left it at the TL;DR version. Now it's just me rambling with a whole bunch of quotes you could put on a postcard. But what exactly am I going to be doing? Well, I'm going to be making making videos online. So if you want to see anything, leave a comment below and I'll do my best to get to you. And don't worry if you've already suggested something. I've got a list as long as my arm of things that— videos I should make because you're all so helpful in suggesting in the comments and other various resources things I should make. And I don't ignore them, trust me. I just, I just put them in a list so I can look at them at a later time. What else is there? Well, Essentially, I've got some income now, I've decreased my burn rate, which means the amount of money it costs for you to live, let's say food, rent, and expenses, to a very small amount. I mean, I don't do much apart from sit in my room and work on my computer and work out, so I don't really have that many expenses. I'm pretty lucky in that way. And so I've got a bunch of cash saved up that I can use to live and then kind of work out this whole business thing. And I mean, the good news is, as I said, I've got some cash coming in for now, but I don't want to rely on that. I want to use that to build something larger. And the idea I kind of have at the moment is something to do with health and nutrition education in terms of an application to do that. But what does that exactly look like? I'm not sure. It would be good if I had a map, but future Dan won't tell me anything at this how this whole venture works out, will ya? Rule number 1 about the future: don't tell people from the past about the future. A known future is already the past. That's how it works. But don't worry, there'll still be plenty of videos around health, technology, and just creative stuff in general. That's kind of the circle I want to— I like to live my life by, is that, uh, learning to create and creating to learn. If you imagine it just goes round and round and round. So that's essentially where I'm trying to go. And I read a quote from a smart person the other day. If you haven't followed him on Twitter already, you should. It's from Naval Ravikant. He's @Naval on Twitter. He posts these beautiful things. Some of the stuff you could, you could, I could get all of his tweets tattooed on my body. I'm not going to do that, right? Because I like tattoos, but not that much. The quote he says is, "You have to keep changing what you do until you become the best in the world at it." And so that's kind of what I'm trying to do. I'm going to keep reinventing what I'm doing until I'm the best person in the world at it. And what's that? Well, I'd like to think it's going to be somewhere in the crossover of the 3 things I've mentioned before: health, technology, and art. So we'll see how it goes. But in the meantime, we'll see you on Sunday. We're doing a livestream. If you have any questions or Q&A, leave them below and I'll get to them on Sunday. But, uh, yeah, I think that's enough. I've rambled enough now. As always, keep learning, keep getting after it, and we'll see you in the next video. Oh wait, if you do want to help out, there's one thing you could do, and that's leave a like or subscribe to the channel or something like that. But otherwise, just being here, just saying hey, that's enough. You don't even have to say hey. I'm rambling too much now. All right, I got to get better at saying goodbyes. Oh, P.S., thank you everyone at Max Kelson. What you did for me is amazing. I love the team there. Love the work there. As I said, was no easy decision. Head was like, this thing is really good that you have right here. But that was like, try it, see how it goes. Worst case, 20 years you look back and you go, oh, I don't regret making that decision. That was just a really challenging time. Peace. Oh my goodness. Oh my goodness. Goodness! Guys, I had an amazing time at the internship today. It's about 7 PM tonight, so I'm home a little late. I was up really early and now it's dark outside, but I'm pumped. You know why? Because when you're learning great things and you're around great people, it's hard not to be full of energy. And I think that's, that's where I get most of my energy from, is from working on hard things, committing to it, doing it, seeing results, and meeting great people.
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


