source: https://aws.amazon.com/certification/certified-machine-learning-specialty/

How I Passed the AWS Certified Machine Learning Exam

6 min readJun 2, 2024

I arrived, I studied, I passed

The AWS Certified Machine Learning — Specialty Exam can be a great way to test your knowledge and consolidate your understanding. The exam can also help build credibility in a competitive field and most importantly, help you develop an industry understanding of Machine Learning.

I would recommend this certification exam to anyone who uses the AWS platform. However, the exam goes beyond AWS and tests knowledge on Machine Learning and Data Science concepts that any seasoned Data Guru should know.

The exam covers 4 domains with varying scoring weights:

Domain 1: Data Engineering (20%)
Domain 2: Exploratory Data Analysis (24%)
Domain 3: Modeling (36%)
Domain 4: Machine Learning Implementation and Operations (20%)

Each one of these domains could be it’s own test, so it’s a lot of information to cover. The exam has 65 questions and is timed to 180 minutes. The exam is proctored and you can take it at home or in a testing center. More information can be found here in the AWS test guide. This article will cover how I passed the exam and any advice I think would be helpful for you.

Gather Your Resources

Every endeavor needs resources to execute it, here are two that I recommend.

Udemy

I used two online resources to study. The first is Udemy’s AWS Certified Machine Learning Specialty 2024 — Hands On! online course that took up the bulk of my study work. I watched every video, took notes on the courses and studied any information I was unfamiliar with. It’s a great, concise course that covers all things Data Science, Machine Learning and AWS. It’s also up to date with current exam trends and emerging Machine Learning technologies. It even includes practice exams after each section and a comprehensive practice exam at the end!

Study Guide

The second resource was an online book, AWS Certified Machine Learning Study Guide: Specialty (MLS-C01). I like the book because it tended to fill in some gaps from Udemy’s online course and having the online book on my kindle app made studying and traveling possible. Udemy is great because it covers general knowledge needed for the exam, but the book looks at Machine Learning through an industry lens and offers real-world advice on how to deploy a model given constraints on time, money and data. The book also has a practice exam at the end of each chapter.

However, the book is worth the read because it includes some helpful resources:

Tips, like when to use CNN-QR vs DeepAR+
Real World Scenarios, like how to effectively deploy a facial recognition model at low cost with limited amount of data.
Notes that can concisely offer more details about a subject

source: https://www.amazon.com/gp/product/B09MBGCRQ7/ref=kinw_myk_ro_title

Get Hands On Practice

You can only read and watch videos for so long, but getting learning into muscle memory is a whole other thing. I highly recommend doing some online courses that offer hands on experience.

Coursera

The course I used is Practical Data Science on the AWS Cloud. The course was originally on Coursera. I liked this course because it included a lab, where you could actually use SageMaker and complete assignments.

SageMaker

If you can’t find hands-on experience within a course then I highly recommend just opening your own AWS SageMaker account and get to work! There are tons of resources that AWS offers, such as notebooks or other documentation.

Practice Tests

To get in the test-taking mind set I encourage taking as many practice exams as possibles. These practice exams can also help you test your knowledge and see where gaps exist. All the resources I listed above have practice exams, but I also used additional exams because I didn’t want to see the same question twice.

The exams from Udemy are great because they break down why each answer is right/wrong and give insights into which area you need to focus more on. I recommend either of these two:

See the exam, Know the exam

This topic will cover general tips for the exam and concepts that I think would be helpful studying:

Almost any service can be integrated with another. Know how they can work together to build an application. Like Polly and Lex to build a chatbot or Kinesis and Rekognition for facial recognition. These are just two examples, but you get the gist.
AWS is very proud of Random Cut Forest, you should probably be familiar with it.
AWS seems to love all things chatbots and Alexa. You should probably be familiar with how to create a chatbot or an Alexa device using AWS services.
XG Boost is very popular right now, it’s best to understand the the model, what it’s used for and how to tune it.
Know all the metrics used for classification inside and out. This includes Precision, Recall, F1, Accuracy, etc. Know the formula, when to use them and the trade offs. This is especially important because these metrics just aren’t useful for the exam, but also useful in your career.
Don’t skip on the hyperparameters. Since there is so much to know, memorizing all the different hyperparameters for each model can seem daunting. However, for each model you should know a few, especially ones that can help with overfitting or underfitting. This is another one I recommend because it’s beneficial to your career.
Know how to problem solve with resources from AWS. From authetican, to data storage to model training and deployment. Review different industry-related problem scenarios and how AWS can be utilized to drive a customer outcome.
Read the question carefully and if you have time, go back and review previously answered questions. This isn’t just an AWS thing, but more of a test-taking thing. Really take your time to understand what is being asked of you.
You won’t know everything, but you should know enough to eliminate answers or make an educated guess.
Skip the information you know, dive into the unknown. Also, not an AWS thing, but more of a test-taking thing. I would skip concepts that I had already mastered and focus on ones that I was still fuzzy on or didn’t already know. This is extremely important for knowledge accumulation.
Find a study partner! It really helped me to have someone that I could talk out these concepts with. I didn’t always remember what I read, but I could almost always remember a conversation I had about a topic.

That’s all I got for now! Leave a comment if this guide helped you passed or if you want to leave your own tips!