View profile

MadeWithML's MLOps Fundamentals

Made With ML
MadeWithML's MLOps Fundamentals
By Goku Mohandas • Issue #1 • View online
All the machine learning fundamentals and MLOps lessons are released! Check out this tweet for a quick summary of all the different topics we’ve covered.
  • 🛠 Project-based
  • 💻 Intuition & application (code)
  • 🏆 26K+ GitHub ⭐️
  • ❤️ 30K+ community
  • ✅ 47 lessons, 100% open-source
Find all the lessons here →

MadeWithML's MLOps Course
MadeWithML's MLOps Course
All lessons released (100% open-source)
We finally completed all the MLOps lessons and we’ve covered everything from product → data → modeling → testing → reproducibility → monitoring and more to really close the loop.
With the help from the community, we’ve become:
But, what we’re most proud of is working with the community feedback to create unique content that follows these core principles:
1. Intuition-first
We never jump straight to code, instead we develop an intuition for the concepts first. This allows us to extend and adapt our understanding of the concepts as this space matures.
2. Hands-on
Instead of just discussing MLOps concepts at a high level, we actually code everything. Take a look at our testing or monitoring lessons to get a sense of the level of technical detail we dive into.
@madewithml is hands down the best MLOps resource that I've come across on the web. Truly added "E" to my ML workflow. @GokuMohandas not just throws concepts at you but goes over whys, hows, tradeoffs, tools & their alternatives via high-quality explanations and code.
3. Engineering
It’s not just about ML, it’s also about adhering to software engineering’s best practices. This includes basic scripting → API design → testing + more.
Satyabrata pal
Recently I have been going through MLOps course by @GokuMohandas

I am completely sold out on the clean code and detailed writeup. This is one of the few ML courses which doesn't stop on just training a model but goes beyond that.
4. Comprehensive
The lessons easily extend to all algorithms, data types (text, image, tabular), frameworks, cloud providers, etc. We avoid choosing any specific tech stack because so many real-world decisions are contextual. However we provide the foundation you need to be able to easily make those decisions and adapt to any current/future tools.
Easily one of the most comprehensive series. I am amazed by how much ground it covers starting with data collection, all the way up to k8s + model monitoring 💯
What's next?
All these lessons will always be 100% open-source and we’ll continue to keep them updated as this space matures. We’ve laid out the foundations needed and we have a lot more content to publish as practices become more widely adopted. We’ll also dive a whole lot deeper into many of the covered topics later this summer.
🙏 Request: Our only ask is that you consider sharing the lessons with fellow coworkers, junior developers and especially new college grads who need to know this content in order to responsibly deliver value with ML.
A bit about me (and why we're doing this)
👋 Connect with me on Twitter and LinkedIn.
Over the past 7 years, I’ve worked on ML and product at a large company (Apple), a startup in the oncology space (Ciitizen) and ran my own startup in the rideshare space (HotSpot). Throughout my journey, I’ve worked with brilliant developers and product managers and learned how to responsibly develop, deploy and iterate on ML systems across various industries.
I currently work closely with early-stage and mid-sized companies in helping them deliver value with ML while diving into the best and bespoke practices of this rapidly evolving space. I want to share this knowledge with the rest of the world so we can accelerate progress in this space.
ML is not a separate industry, instead, it’s a powerful way of thinking about data, so let’s make sure we have a solid foundation before we start changing the world. Made With ML is our medium to catalyze this goal and though we’re off to great start, we still have a long way to go.
A bit about our newsletter
  • We take your inbox very seriously so this newsletter is reserved for major releases only but follow us on Twitter and LinkedIn for more frequent updates.
  • We recently switched our email marketing client after we discovered that a significant portion of our audience was not receiving our monthly newsletters. Hopefully our email deliverability improves with this new service.
  • Many of you signed up for our newsletter sometime over the last three years. Since then, we’ve gone through a name change (practicalAI → Made With ML) and a drastic pivot.
Did you enjoy this issue?
Goku Mohandas

Learn how to responsibly deliver value with machine learning.

In order to unsubscribe, click here.
If you were forwarded this newsletter and you like it, you can subscribe here.
Powered by Revue