HOW I PASSED MY AWS MACHINE LEARNING SPECIALTY EXAM….

Hi everyone! I’m Olalekan Fagbuyi, and I’m excited to share that I recently passed the AWS Machine Learning Specialty Exam. Since then, I’ve received numerous requests about the study materials and resources I used to prepare. To help others on their journey, I decided to put together this comprehensive guide that outlines the key materials and strategies that were instrumental in my success.

I have organized this post into the following sections:

1. Required Past Knowledge: Understanding the foundational concepts and prerequisites that will set you up for success.
2. Reading Material: The essential books, articles, and documentation that provide a solid theoretical background and practical insights.
3. Practice Material: The best resources for hands-on practice and mock exams to test your knowledge and readiness.
4. Final Thoughts: My personal tips, experiences, and recommendations to help you navigate the preparation process and ace the exam.
5. Required Past Knowledge

AWS recommends the following prerequisites to pass this exam:

• 1 to 2 years of experience developing, architecting, or running machine learning and deep learning workloads on the AWS Cloud.
• Understanding and intuition behind basic ML algorithms and experience performing basic hyperparameter optimization.
• Experience with machine learning and deep learning frameworks.
• The ability to follow model-training, deployment, and operational best practices.
• Prior experience with AWS, in particular streaming services (Kinesis family) and batch services (SageMaker, AWS Glue, Batch, Data Pipelines)

Prior to the exam, I had the following data science / cloud related experience:

• 5+ years of hands-on experience working with Python, SQL, and other software tools to manipulate and analyze data.
• 2 years of self-study as a data scientist.
• 9 months into a 12-month Management Analytics Master’s program
• 6 months of experience using the AWS SageMaker console and other AWS cloud services to develop and run ML workloads
• I passed the AWS Certified Cloud Practitioner exam, which sets the foundation for understanding AWS.

Having said that, you do not need a master’s degree to pass this exam. Same with coding, you will NOT code, nor do you need to do code reviews, to be ready for this exam. However, those skills could be of help.

2. Reading Material

I spent 6 weeks preparing for this exam. I dedicated the first 4 weeks to studying the materials listed in this section, and the final 2 weeks to focusing on the practice exams listed in the next section.

I. Exam Guide. This 11-page document breaks down the structure, areas of focus, scoring, cost, and other relevant information required for the exam. I strongly recommend that this be the starting point for anyone interested in the exam.

II. AWS ML Specialty Course: This Udemy course, taught by Frank Kane, a nine-year Amazon machine learning expert, and Stephane Maarek, an AWS expert, includes a 9-hour video course, a 30-minute quick assessment practice exam, and four hands-on labs. The course covers topics like model tuning, feature engineering, and data engineering and provides valuable experience for those preparing for the real exam.

III. AWS Machine Learning Plan: This learning path on the AWS Skill Builder website contains hands-on labs, interactive modules, and detailed tutorials covering the full spectrum of machine learning topics relevant to the exam. One of the standout features is the “Exam Readiness” section, which specifically focuses on equipping candidates with the knowledge and skills necessary to excel in the certification exam. This section provides practice questions, exam strategies, and a thorough review of the key concepts and services, such as SageMaker and AWS’s suite of AI services.

IV. AWS SageMaker Documentation: This is a comprehensive resource that offers over 6,000 pages of documentation on SageMaker’s features, including data preparation, model training, deployment, and monitoring, which are crucial for the exam. I doubt anyone reads the entire documentation, but it’s advisable to consult it whenever you find yourself stuck on certain concepts or identify your weak areas during the practice exams. This is what I did to get a better understanding of the MLOps sections.

V. AWS AI Services: Understanding AI services like Translate, Transcribe, Lex, Polly, and others on the AWS website is instrumental in the exam. These services are key components in the AWS ecosystem, providing solutions for natural language processing, speech recognition, text-to-speech, and language translation. I spent a few hours looking into the functions and use cases of each of these services on the AWS AI website.

VI. SageMaker Technical Deep Dives: These are 16 YouTube videos of approximately 5 hours presented by Emily Webber, a principal ML specialist at AWS. These videos provide in-depth explanations and practical demonstrations of SageMaker’s capabilities, covering everything from data preparation and model training to deployment and monitoring. Emily Webber’s clear and concise teaching style makes complex concepts accessible, and her use of real-world examples helps solidify understanding.

3. Practice Material

One of the most crucial steps in this process is dedicating ample time to practice material. This step is non-negotiable and serves as the foundation for ensuring success. By committing to regular practice exams, you not only reinforce your knowledge but also develop the skills and confidence needed to excel on the actual test. Here are some of the practice tests I took.

I. AWS Free Practice: I used three free practice materials from AWS. They are a good starting point because they follow the format of the actual exam, though the questions are slightly easier than the real thing.

• Exam Prep Official Practice Exam Set: 20 Questions
• AWS ML Specialty Sample Exam Questions: 10 Questions
• AWS Exam Readiness : 43 Questions. This is available in the AWS Machine Learning Plan module mentioned earlier.

II. AWS ML Specialty Exam Practice by Frank Kane: 75 Questions (10 Warm-up + 65 Exam Questions)

III. AWS ML Specialty Exam Practice by Abishek Singh :140 Questions (10 Warm-up + 2 * 65 Exam Questions)

IV. AWS ML Specialty Exam Practice by Neal Davis and Karim Elkobrossy: 120 Questions (Split into 6 blocks of 20 Questions each)

V. AWS ML Specialty Practice Exam by Jon Bonso: 158 Questions (2 * 65 Exam Questions + 28 Bonus Questions)

II to IV are available on Udemy, while V is available on the Tutorial Dojo’s website. I attempted each practice exam three times, with my scores going from 58% at the beginning to 88% at the end. I eventually scored in the early 80s on the exam.

The practice exams offered by Tutorial Dojo closely resembled the actual exams. It also has a feature that enables users to focus on specific areas; for example, I was able to focus on questions from the MLOps and Data Engineering sections to spot my weaknesses, which I later improved on by reading the SageMaker documentation and other material.

4. Final Thoughts

The AWS Machine Learning Specialty exam is challenging but achievable with the right preparation and resources. It is crucial to have a solid understanding of machine learning concepts and AWS services. The exam tests your ability to apply this knowledge to real-world scenarios, so hands-on experience with AWS SageMaker and other ML services is highly beneficial.

Remember that the exam is not just about memorizing facts but also about understanding how different services work together and how to choose the right service for a given task. Therefore, it is essential to practice with various scenarios and use cases to solidify your understanding.

Finally, don’t underestimate the importance of time management and practice exams. Create a study schedule that works for you and stick to it. Take as many practice exams as possible to familiarize yourself with the exam format and identify your weak areas.

I hope this guide has been helpful in your preparation for the AWS Machine Learning Specialty exam. If you have any questions or comments, please feel free to leave them below. Good luck on your exam!