Hi, I’m Logan 👋 Machine learning software engineer, optimist, and father of 5.
I obsess with helping others (especially engineers) understand AI systems so they can work well for everyone. I share this via:
My writing:
A free ML road map:
I turned 27 today and I’m loving life.
I’ve got an awesome wife and 5 amazing kids (including two sets of identical twins!). I get to work on my passion every day and share it with all of you.
I’m super blessed. I thank you all for your support in my endeavors.
I guess this is what happens when you correct an incorrect (and harmful) statement made by a larger account.
It's a shame because I really enjoyed most discussions with her.
MLX is a BIG deal for the ML community. It significantly lowers the barrier of entry for ML research and makes running models locally much easier.
I wrote a step-by-step guide to training your first ML model on your Mac in less than 5 minutes. Link below 👇
Does anyone else find it interesting that the engineers creating paid courses teaching software engineering often aren't actually working in a software engineering role?
The Machine Learning Road Map is here! 🥳It's your guide to navigate the best machine learning resources for FREE. Link in my profile and below.
I'll be spending this month going through all the resources myself to improve it.
Here are the resources I'm most excited about:
A big thing I see a lot here: people pretending to be experts when they’re just starting to learn something.
I don’t know why anyone feels the need to do this, but you’ll get a lot further and make a lot more connections by being honest.
Personal update: the boys DNA test shows they're identical. That's two sets of identical twins for us
The odds of that happening: 1/277,778
The odds of being struck by lightning in your lifetime: 1/15,300
I'm staying inside during storms
I’m excited to announce the ML Road Map-Turbo! It’s a streamlined road map to help anyone get up to speed on ML fundamentals quickly. It makes sense of the many high-quality ML educational resources available.
Here’s a walkthrough of what's included:
I've really enjoyed getting to know a lot of frontend developers on X, but you guys limit yourselves so much.
Why label yourself as a React dev? Or even a Nextjs dev? You don't think you can use other frameworks?
@bindureddy
You clearly doesn't understand the drug development process. The work these models can do in minutes would have previously taken hours using other computational methods.
I wish anyone would invent a drug that cures cancer or diabetes.
MLX team needs a raise.
I’m going through my list of ML updates (96 total) from the past two weeks and MLX has 7 entries I think need to be shared. This is after trimming it down to about half. Their pace is incredible.
"Tech is easy money" really bothers me. There are VERY few devs doing little work and making >$200k/yr. This saying sets so many people up for disappointment.
Can anyone learn how to code? Yes.
Will most people enjoy it? Probably not.
Has anyone else debated cancelling their other streaming subscriptions and just watching YouTube full time?
YouTube has enabled human creativity more than anything else in history and I don't think it even comes close.
Tired of the doomsayers saying AI will take all our jobs.
The reality: we’re not even close to meeting the demand of people needed for machine learning and the field is growing VERY quickly.
Your social media feed should be a source of inspiration, not exhaustion.
If social media tires you out, take a step back and reconsider the way you're using the platform.
@AravSrinivas
All the people I’ve met at work are huge nerds and it’s celebrated.
But that doesn’t mean they have to desire to build things all the time.
Keep in mind that very few people are AI experts.
I've worked as an ML engineer at Google and Microsoft and as an AI research scientist and I'm not even close to being an expert.
Claiming to be an expert and sharing nonsense harms everyone's understanding of AI.
X has changed me from wasting time on the internet to being more productive.
I get way more out of the conversations I have with all of you than I did from mindlessly scrolling Reddit.
Just a friendly reminder that being kind and supportive is a super easy way to build people up.
It takes almost no effort on your part and makes a huge difference to others.
@bindureddy
It’s also an early-stage novel device so I highly doubt they expected mass adoption.
We’ll have to see what happens when they get the price down.
Tech is about people.
Whether it's the people you build for or the people you build with, it's the people that make it worth it.
X has a done a great job of reminding me about that.
I want to help people become better but I also want to help people feel better about themselves.
Here's my story:
I grew up in a great area with a great family. I was a good athlete, had high intellect, and had a lot of friends. I never really faced any serious struggles. My
There’s too much of a focus on hustle culture on X.
Messages like “you don’t grow on X because you don’t work hard enough” are only posted for impressions and contribute to the toxicity I see here.
Get them off my feed.
There seem to be two types of indie devs:
Those building a portfolio to get a job at a great company
And those building because they never want to work for a company
I'm convinced the key to life is finding your people. People who share your interests and appreciate who you are.
X has been awesome for finding my people. Thank you everyone! 🙂
Stanford's Intro to Machine Learning course (CS229) is a perfect example of understanding basic topics before getting into something more complex.
It takes half the course before they get into neural networks. Most new machine learners jump right into them.
After 18 weeks on paternity leave, I’m headed back to Google today.
I loved the extra time I had with my kids, but I’m excited to get back to building cool things. 👍
I've decided to create a series of posts on my blog about developing the soft skills to succeed as a software engineer.
I was debating an ebook of some sort, but this way I can things out piece by piece and it's much more accessible. Let me know what you think.
Networking is reciprocal. If you aren't willing to help others, why should they help you?
Instead of taking time to grow your network, take time to help your current network.
If you're ever feeling obsessed with your follower count, take a step back to focus on who you're following instead. It's the more important number and it's overlooked.
Curate your feed, curate your mind.
The terms small language model (SLM) and large language model (LLM) are funny to me.
We know they're only getting bigger. By next year today's LLMs will be considered SLMs. 😂
The Machine Learning Road Map focuses so much on CS229 if you want to work in ML because it teaches the skills companies are looking for.
I've annotated and timestamped all the topics to make it easier for you to learn them. Check out the entire list of topics below:
The best piece of career advice I have ever received was to stop worrying about money.
The area you work in, the product you work on, and the people you work with have a much greater effect on your career growth and happiness. Optimizing these will pay dividends in the long run.
The thing I love most about software: as I learn more, I realize just how little I know.
A lot of that has come from conversations with all of you. Thank you!
Being an excellent developer is more than just having domain knowledge. You have to think like an engineer.
Every great software engineer I know is obsessed with understanding how things work. They simply have to know.
It took me a while to realize that networking is just building a personal brand and it's always happening.
Being genuine is what's helped me do this best. What do you do to build your personal brand?
I love watching videos from YouTubers who speak English as a second language.
They enunciate so much better and their communication is much clearer than English speakers. As someone who solely speaks English, I appreciate it.
Rabbit holes are the reason I love tech.
Find something interesting? Dig into it.
There are always new things to learn and chances are that knowledge will help you down the line.
If you're getting in AI, understand the engineering behind the machine learning.
We're going to see ML engineering jobs 𝗲𝘅𝗽𝗹𝗼𝗱𝗲. Training, experimenting, and serving models will only get more important.
People won't think of working in AI primarily as research anymore.
I've been sharing things online for about 9 months now and the biggest thing I've realized is we need more people in the machine learning community online to fight the misinformation about ML spread by those who don't understand it.
My goal for 2024 is to put all my data (notes, ideas, tasks, projects) into a second brain that can easily be made available to an LLM.
By the end of 2024, I'll have a second me to assist with everything I need to get done.
Then I'll make it easy for anyone to set up.
Here are four simple steps to start learning ML:
1. Learn Python
2. Learn linear algebra
3. Watch the CS229 lectures on YouTube
4. Build models in Kaggle Competitions
If you have any questions, shoot me a DM.
It's crazy that X gives me direct access to machine learning experts.
I frequently ask questions and get answers within minutes. No other platform provides this.
I just realized the ML Road Map is at 74 stars! I'll be working on it this upcoming week to add more resources and streamline it so anyone can learn ML for free!
If you want to learn, shoot me a DM.
I'm convinced the two most important things to become a better software engineer are completely separate from technology. They are:
1. Sleep
2. Exercise
I can immediately feel my throughput decrease without them.
@catalinmpit
Woohoo! Welcome to the club! You’ll have to let us all know where you place lights/camera, etc for filming videos. I’ve found fitting them around the larger screen to be a bit clunky.