Sivadharan K
5 min readJun 11, 2022

AWS Certified Machine Learning — Specialty (MLS-C01) — Some tips to pass the exam

In May 2022, I passed the Machine Learning Specialty Certification (MLS-C01) by Amazon Web Services (AWS). In this article, I have curated some strategies and tips that will be helpful for someone considering taking this exam.

It is an ‘Advanced’ level AWS certification that mentions “Earners of this certification have an in-depth understanding of AWS machine learning (ML) services. They demonstrated ability to build, train, tune, and deploy ML models using the AWS Cloud. Badge owners can derive insight from AWS ML services using either pretrained models or custom models built from open-source frameworks.”

It validates our general ML skills and also based on use cases covering end to end ML pipeline with respect to AWS infrastructure.

1. Exam Prep Resources

I took the Udemy course “AWS Certified Machine Learning Specialty 2022 — Hands On!” by Frank Kane and Stephane Maarek. With this course, I was able to set a clear boundary of what I need to exactly learn.

I also took another Udemy course “AWS Certified Machine Learning Specialty (MLS-C01)” by Chandra Lingam. With this course, I got more insights and better understanding. This course material has been curated with multiple references, along with a practice exam with explanations of answers.

I have also gone through the book “AWS Certified Machine Learning Specialty: MLS-C01 Certification Guide” by Somanath Nanda and Weslley Moura. This book covers several important topics about the foundations of ML and diving into certification-related topics. This book is well designed for Topics — “Data Engineering, EDA, Modelling with Sagemaker, High Level AI services”. But the ‘ML Implementation and Operation’ is not covered as a separate topic, instead it has been intermingled as part of other topics.

I do not have any conclusions like which course is best among all. I would suggest you to go through the review and do some preview to find which course suits you the most. But choosing single course is also fine, I think.

This exam has been categorized as ‘Difficult’ by many people. Yes, it was bit challenging but totally do-able if you are well prepared. Going through the course material(s), will help to get baseline score (around 75 to 80%).

Boosting up the score further depends on how we comprehend the questions and answers wisely and how much we go through the SageMaker document. For example, I have got a question on ‘Measure Pretraining Bias’. This topic is available in SageMaker document , but not in any course material that I read through. We have to prioritize and go through some important SageMaker document pages at least. I have listed out some URLs as part of this article that I came across.

2. Practice Exam

The practice tests would help you to know what kind of questions and from which perspectives your knowledge will be tested in the real exam and also to validate your exam readiness before going for actual exam.

You can find some of them on Udemy. I used the below Practice Exams.

  • ‘AWS Certified Machine Learning Specialty Full Practice Exam’ by Frank Kane.
  • ‘AWS Certified Machine Learning Specialty: 3 PRACTICE EXAMS’ by Abhishek Singh.
  • ‘AWS Certified Machine Learning Specialty Practice Exams’ by Jon Bonso.
  • Udemy course “AWS Certified Machine Learning Specialty (MLS-C01)” by Chandra Lingam. This course also includes Practice Exams.

3. Exam Complexity

  • During practice exams, I took around 1 to 1.5 hours to complete the test. But the real exam took around 2.5 hours to complete. Many of the use cases and the answers are bit lengthier when comparing to the practice exams. So it took more time to comprehend. The positive side is that use cases with lengthy sentence have more key words that will help to narrow down the correct answers more precisely.
  • There may be some in-depth questions in general Machine Learning and also from SageMaker document. As per AWS document “Each exam also contains a number of unscored questions that don’t count toward your raw score. The unscored items are newly developed questions that are being evaluated. Because they are new, we don’t have enough statistical data to gauge their difficulty.”. I would suggest to give focus on doing base level preparation (Exam Prep Resources) and then going through important SageMaker document pages for extra preparation. This approach will bring you a decent score.

4. Some important URLs

https://docs.aws.amazon.com/sagemaker/latest/dg/algos.html
https://docs.aws.amazon.com/sagemaker/latest/dg/IC-HowItWorks.html
https://docs.aws.amazon.com/sagemaker/latest/dg/your-algorithms-training-algo-dockerfile.html
https://docs.aws.amazon.com/sagemaker/latest/dg/model-ab-testing.html
https://docs.aws.amazon.com/sagemaker/latest/dg/seq-2-seq-howitworks.html
https://docs.aws.amazon.com/sagemaker/latest/dg/notebook-interface-endpoint.html#notebook-private-link-restrict
https://docs.aws.amazon.com/sagemaker/latest/dg/model-ab-testing.html#model-testing-target-variant
https://docs.aws.amazon.com/sagemaker/latest/dg/nbi-software-updates.html
https://docs.aws.amazon.com/sagemaker/latest/dg/model-monitor-model-quality-baseline.html
https://aws.amazon.com/blogs/machine-learning/millennium-management-secure-machine-learning-using-amazon-sagemaker/
https://aws.amazon.com/blogs/machine-learning/configuring-autoscaling-inference-endpoints-in-amazon-sagemaker/
https://aws.amazon.com/blogs/machine-learning/ml-explainability-with-amazon-sagemaker-debugger/
https://aws.amazon.com/blogs/machine-learning/private-package-installation-in-amazon-sagemaker-running-in-internet-free-mode/
https://aws.amazon.com/blogs/machine-learning/access-amazon-s3-data-managed-by-aws-glue-data-catalog-from-amazon-sagemaker-notebooks/
https://aws.amazon.com/blogs/big-data/how-to-access-and-analyze-on-premises-data-stores-using-aws-glue/
https://aws.amazon.com/blogs/security/secure-deployment-of-amazon-sagemaker-resources/
https://aws.amazon.com/about-aws/whats-new/2019/01/amazon-sagemaker-batch-transform-now-supports-tfrecord-format/
https://docs.amazonaws.cn/en_us/sagemaker/latest/dg/apache-spark.html
https://docs.aws.amazon.com/forecast/latest/dg/dataset-import-guidelines-troubleshooting.html
https://docs.aws.amazon.com/lex/latest/dg/howitworks-custom-slots.html
https://docs.aws.amazon.com/glue/latest/dg/dev-endpoint-tutorial-sage.html
https://docs.aws.amazon.com/forecast/latest/dg/aws-forecast-algo-cnnqr.html
https://docs.aws.amazon.com/sagemaker/latest/dg/experiments.html
https://docs.aws.amazon.com/personalize/latest/dg/recording-events.html
https://towardsdatascience.com/natural-language-processing-count-vectorization-with-scikit-learn-e7804269bb5e
https://corporatefinanceinstitute.com/resources/knowledge/other/positively-skewed-distribution/
https://github.com/awsdocs/amazon-sagemaker-developer-guide/blob/master/doc_source/your-algorithms-training-algo-dockerfile.md
https://www.quora.com/Whats-the-difference-between-a-naive-Bayes-classifier-and-a-Bayesian-network?share=1
https://towardsdatascience.com/introduction-to-word-embeddings-4cf857b12edc

5. Use Medium

As like my other AWS exam, I have been immensely benefited from the articles shared in Medium by those already taken the exam. Here are some URLs to refer.
1. https://medium.com/@darya_petrashka/aws-machine-learning-specialty-you-can-do-it-e2859b0f0407?source=user_profile---------0-------------------------------
2. https://medium.com/swlh/cheat-sheet-for-aws-ml-specialty-certification-e8f9c88566ba
3. https://towardsaws.com/how-i-passed-the-aws-machine-learning-specialty-certification-4b5e028b1dd3
4. https://dipayan-x-das.medium.com/journey-towards-aws-certified-machine-learning-specialty-mls-c01-certification-6272a4ef5423
5. https://towardsdatascience.com/advice-tips-for-passing-aws-machine-learning-specialty-102f7f5f99a0
6. https://javier-ramos.medium.com/aws-machine-learning-certification-exam-tips-2a7679a83e73

Thank you for reading. Good luck!