Building Machine Learning Systems That Don't Suck 11/08/24, 11:17 AM

"This is the best machine learning course I've done.

Worth every cent."
— Jose Reyes, AI/ML at Cevo Australia

Building Machine
Learning Systems That
Don't Suck
A live, interactive program that'll help you build production-ready
machine learning systems from the ground up.

Next cohort: November 4 - 21, 2024

Check the schedule for more details about upcoming cohorts.

I want to join! Sign in

Learn how to design, build, deploy, and scale

machine learning systems to solve real-world
problems.
https://www.ml.school/ Page 1 of 13
Building Machine Learning Systems That Don't Suck 11/08/24, 11:17 AM

I'll lose my mind if I see another book or course teaching people the
same basic ideas for the hundredth time. Most people are stuck in
beginner mode, and finding help to solve real-world problems is
hard.

I want to change that.

I started writing software 30 years ago. I've written pipelines and

trained models for some of the largest companies in the world. I
want to show you how to do the same.

This is the class I wish I had taken when I started.

This program will help you unlearn what you think machine learning
is. It's a practical, hands-on class where you'll learn from years of
experience and real-world examples.

When you join, you get lifetime access to the following:

18 hours of live, interactive sessions. We'll use this time to

discuss the first principles behind building machine learning
systems.

10 hours of step-by-step coding instructions. These practical

sessions will show you how to build an end-to-end system from
scratch.

A final project where you'll build a complete solution and

receive direct feedback on your work.

100 coding assignments and practice questions.

The entire source code of a working production system. It's

yours. You can change and use it as you see fit.

https://www.ml.school/ Page 2 of 13
Building Machine Learning Systems That Don't Suck 11/08/24, 11:17 AM

A private community where you'll collaborate with thousands of

people from different backgrounds.

Direct access to your instructor.

Lifetime access to every past and future cohort.

Program certificate upon completion.

And the best part is that you only pay once to join. There are no
monthly fees. No annual fees. No hidden costs. You pay once to join
and benefit forever until the end of time.

The program won't be easy. It'll take time and effort. But if you want
to use machine learning to solve real-world problems, this is the
class you don't want to miss.

Who Is This Program For?

This is a practical, hands-on program for
technical professionals who are ready to put in
the work.
This program is for software engineers, data scientists, data
analysts, machine learning engineers, technical managers, and
anyone anyone who wants to use machine learning to solve real-
world problems.

Here are the criteria to succeed in the program:

You have experience writing code. We'll use Python throughout

https://www.ml.school/ Page 3 of 13
Building Machine Learning Systems That Don't Suck 11/08/24, 11:17 AM

the class, but you won't have any problems if you know any
other language.

You are familiar with basic machine learning terminology. This is

not an introductory class. We'll move quickly over the basics to
focus on the fundamental ideas that make systems work.

You are ready to put in the work to succeed.

"I have learned a ton from Santiago in his class and it was actually what helped
inspire me and get into the MLOps work that I'm doing now. Truly one of the most
helpful online courses for doing real, full-scale machine learning."

Brian H. Hough
Software Engineer

What Will You Learn?

You'll come out with practical skills and
insights into what it takes to build systems that
work in the real world.
Here is a summary of what makes this program unique:

You'll design and write the code to build an end-to-end

machine learning system starting from scratch.
https://www.ml.school/ Page 4 of 13
Building Machine Learning Systems That Don't Suck 11/08/24, 11:17 AM

machine learning system starting from scratch.

You'll learn best practices to tackle the most significant

challenges machine learning engineers face to build, evaluate,
run, monitor, and maintain machine learning systems in real-
world scenarios.

You'll learn how to use techniques like active learning,

distributed training, adversarial validation, human-in-the-loop
deployments, model compression, test-time augmentation,
testing in production, among many others.

You'll learn how to create training, deploying, monitoring, and

inference pipelines using Amazon SageMaker and open-source
tools.

Forget about theoretical concepts. This program will show you some
of the things I've learned from real-life examples I've built during
more than 30 years in the industry.

Check the program syllabus →

https://www.ml.school/ Page 5 of 13
Building Machine Learning Systems That Don't Suck 11/08/24, 11:17 AM

Real-life examples Live, interactive Code

and case studies sessions walkthroughs
Learn from practical Ask questions and Step by step coding
experience building interact with the instructions to help you
machine learning instructor and other build a production
systems that work in students in real time. system from scratch.
the real world.

Upcoming Schedule
Every iteration of the program gives you 18
hours of hands-on, live training spread over 3
weeks.
Here are the upcoming cohorts:

Cohort 16: November 4 - November 21, 2024. 1:00 PM EST

Live sessions will take place every Monday and Thursday at the
same time. On Wednesdays, we'll host office hours when you can
bring your questions projects or anything else you want to discuss.

Monday: Live session. 2 hours.

Tuesday: Individual work.

Wednesday: Optional office hours.

https://www.ml.school/ Page 6 of 13
Building Machine Learning Systems That Don't Suck 11/08/24, 11:17 AM

Thursday: Live session. 2 hours.

Friday: Individual work.

Do not wait for a specific cohort to join the program. You have
lifetime access to every past and future cohort, and the sooner you
join, the more time you have to prepare.

Every session is recorded. You can attend live or watch the recorded
version later.

"This is one of the best classes I've ever purchased over the internet. Santiago is
a terrific teacher. The ability he has to share knowledge is fantastic. I recommend
this course. Worth 10x what he's charging."

Sal DiStefano

Ready To Join The Program?

You'll get lifetime access. No monthly fees. No
annual fees. No hidden costs.

https://www.ml.school/ Page 7 of 13
Building Machine Learning Systems That Don't Suck 11/08/24, 11:17 AM

$500
$300
Pay once. Access forever.
Pay once to join the program and get lifetime access. You can
participate in as many iterations as you'd like. No restrictions.

Join now

Enjoy 18 hours of live, interactive sessions

Watch 10 hours of step-by-step coding instructions

Practice with 100 coding assignments

Access the complete source code of a production system

Learn how to start freelancing on Upwork

Get feedback and support from the community

Get direct feedback from your instructor

Program Syllabus
This program will teach you the practical skills
and insights that will help you build machine
learning systems.
https://www.ml.school/ Page 8 of 13
Building Machine Learning Systems That Don't Suck 11/08/24, 11:17 AM

Here are the contents of the six live sessions of the program:

Session 1 - How To Start (Almost) Any Project

What makes production machine learning different from what
you've learned.

The strategy to solve the right problem using the right solution.

Critical questions to ask before starting any project.

Problem framing, inversion, and the haystack principle for

building successful applications.

The first rule of machine learning engineering and how to start

building.
Data collection strategies. A technique to determine how much
data you need.

The problem of selection bias and how to deal with it.

Labeling data. Human annotations, natural labels and weak

supervision.

Active learning using the uncertainty and diversity sampling

strategies.

Session 2 - How to Build a Model

The role of data cleaning and feature engineering to build
better models.

Turning data into numbers using vectorization techniques.

Producing homogeneous features using normalization and

standardization.

Handling and interpreting missing values using imputation

https://www.ml.school/ Page 9 of 13
Building Machine Learning Systems That Don't Suck 11/08/24, 11:17 AM

Handling and interpreting missing values using imputation

techniques.

The approach to choosing the best model to solve any

problem.

Random baselines and the zero-rule algorithm.

How to use overfitting to build models that don't suck.

Hyperparameter tuning and experiment tracking.

Measuring the quality of your holdout set.

An introduction to distributed training using data parallelism

and model parallelism.

See the remaining sessions

"This is an awesome course! This is my second round and I continue learning. I

recommend it with complete confidence."

Juan Olano
Machine Learning Engineer

Frequently Asked Questions

If you can't find the answer to your question,
please reach out on social media and I'll be
https://www.ml.school/ Page 10 of 13
Building Machine Learning Systems That Don't Suck 11/08/24, 11:17 AM

please reach out on social media and I'll be

happy to help.

How long will it take to complete the program?

If you are attending the live sessions, you should set aside a
minimum of 4 hours every week during the three weeks of the
program. This commitment will be enough for engineer leaders or
anyone not interested in the coding portion of the program.

Those interested in implementing the concepts discussed in class

should set aside 2 to 4 hours weekly to complete the code
walkthroughs and work on the assignments.

Are live sessions recorded?

Yes, we record every live session. You can decide when to attend
classes live or catch up asynchronously later using the recording.

I'm a complete beginner. Is this program helpful for me?

This program is not an introduction to machine learning.

While we'll discuss many fundamental ideas behind machine

learning, beginners will find the sessions go much faster than what's
optimal for them.

What does "lifetime access" mean?

You only pay once to join the program and get immediate access to
every past, present, and future cohort.

Every new iteration of the program is better than the ones before.
Many students take classes once and then join a later cohort to
https://www.ml.school/ Page 11 of 13
Building Machine Learning Systems That Don't Suck 11/08/24, 11:17 AM

Many students take classes once and then join a later cohort to
benefit from the updates.

The lifetime access removes any pressure from having to complete

the program when life gets in the way.

Hey! I'm
Santiago.
I'm the instructor of the program.
I'm a machine learning engineer with over two decades
of experience building and scaling enterprise software
and machine learning systems.

I love neural networks. I love to make them work at

scale.

From 2009 to 2023, I built products for Disney, Boston

Dynamics, IBM, Dell, G4S, Anheuser-Busch, and NextEra
Energy, among other clients. I learned about trade-offs
and how to create products that work.

I started this program in March 2023. Since then,

thousands of students have graduated, and I can't wait
to meet you in class.

https://www.ml.school/ Page 12 of 13
Building Machine Learning Systems That Don't Suck 11/08/24, 11:17 AM

Copyright © 2024 Tideily LLC

All rights reserved.

https://www.ml.school/ Page 13 of 13