The Path to Clear AWS Machine Learning Exam
I just passed the exam on 10/31. I have read many blogs about passing this exam, but none of them spelled out the exact paths to pass this exam from 0 to 60 based on my experience. If this is what you are looking for, you have come to the right place. But here is my relevant background that helped me pass this exam. I have nine years of working in the AWS environment, four years in Data Analytics/ML software development, and a Harvard Data Science certificate before taking this exam.
- You need some working experience with the AWS environment. I recommended focusing on these services: EMR/Spark cluster, EC2 spot/reserve instance, EBS, S3, SQS/SNS, Lambda, and how to encrypt data at rest for S3. If you can ace the AWS cloud practitioner certificate, then you are ready.
- Know Python2/3. If you must pick between R, Julia, Python, stick with Python. There are plenty of Python courses on Udemy.com. Select the Python course’s highest rating and wait for the sale, which happens very regularly between $12.99 and $9.99. Black Friday usually has the sale for $9.99/course. Great deal. Use Jupyter Notebook to write Python code; see here https://jupyter.org/ for setup. I also highly recommend using a full-blown Jupiter IDE (my preference) that you can debug code directly in the notebook, great for Python beginners, see here https://www.jetbrains.com/dataspell/. The beta is free for now.
- Write some Python, ML/Deep Learning codes. You only need to pick one of the two paths below.
If you like math, theories, and intuitions, I recommend this route. It may take longer than Kaggle courses below.
Skip the SQL, Geospatial, Ethics, Explainability unless you want to learn the topics.
4. Ok, now you are ready to prep for the exam. If your company provided acloudguru.com for free, use the AWS ML course and the practice exam on the acloudguru site. These courses below were what I used to pass the exam, and I listed them in the order of consumption.
a. Frank and Stephane are both great presenters. I bought other courses from them before, and they did not disappoint. https://www.udemy.com/course/aws-machine-learning/learn/lecture/16397700?start=15#overview. After you watch all the videos, download the course slide and use it for quick review and recall. Don’t buy Frank exam, he has only one full exam.
b. https://explore.skillbuilder.aws/learn/course/27/play/54/exam-readiness-aws-certified-machine-learning-specialty. Watch the videos and do all the exam questions. Great training video and 35 questions exam from AWS and it is free.
c. https://www.udemy.com/course/aws-certified-machine-learning-specialty-full-practice-exams/learn/quiz/4755096/results?expanded=606046360#overview. Do these exams first as they are a little easier to build some exam toughness before the next one. Review all the questions, don’t memorize, read all correct and wrong explanations and ingest the concepts. Navigate to the link in the explanation for more detail background.
d. https://www.whizlabs.com/learn/course/aws-certified-machine-learning-specialty/281/pt. Do the same, review all answers and check out the background reference links. This is a very hard practice exam, harder than the real exam, and if you can get close to 70% correct, you are ready.
5. Handy tips and study notes.
On the final note, it is crucial to work through the questions quickly and efficiently, or you will run out of time. I found that by searching out the question immediately (without reading the entire question) and looking at the answers, you should be able to pick out the correct answer by the process of elimination. I only read the whole question carefully when there are two or more plausible choices as answers. Good luck with the exam!
Kiet Ly is the Chief Data Architect at E-INFOSOL where he is currently helping federal customers migrate big data analytics on AWS. Kiet has extensive experience architecting and building AWS data analytics applications in financial services and defense.
Interesting Fact About Kiet: As a child he was stranded on a remote island for several days!