AWS Training and Certification Blog

Building your machine learning skills from zero

Navigating the tech landscape without a traditional computer science background has been a challenging, yet rewarding, journey for me. Approximately two years ago, I was preparing to take the AWS Certified Cloud Practitioner exam, and if you had asked me if I could ever imagine working with distinguished technologists and some of the largest enterprise companies, I would’ve laughed in your face.

In this blog, I’ll share my non-traditional journey to a career in tech and how this path led me to explore the intricate world of machine learning (ML). I’ll also share advice and resources for learners wanting to make the leap.

About me

Jenny Dassas

I’m a former, failed chemical engineering student turned MBA who spent the majority of my career working in the public sector. When I learned about AWS re/Start, a 12-week intensive program that teaches cloud computing skills, I figured I’d give it a shot. Although it wasn’t chemistry, cloud computing sounded like an exciting pivot as I’ve always been fascinated by the various domains of technology. Through the program, I gained hands-on experience with AWS Cloud services and developed skills in Linux, Python, database design, and more.

Through my hard work in the program, I earned a spot in AWS Tech U, a 48-week accelerated workforce development program consisting of on-the-job training and project-based learning. My capstone project allowed me to get my first taste of ML by leveraging Amazon Transcribe, Amazon Comprehend, and Amazon Rekognition to analyze speech and facial expressions. This internal tool provided constructive feedback for AWS Certified speakers. Although at the time I was only able to engage with ML services at the application layer, this experience sparked my interest in further developing my skills. My confidence with grasping advanced data science concepts was modest but through determination and continued learning, I knew I could expand my capabilities.

In my role as a Customer Solutions Manager at AWS, I am responsible for helping customers adopt our tools and technologies. For my project-based learning, I was assigned to work with an Enterprise Resource Planning company on an Alexa Smart Properties project. While shadowing a customer call, the Senior Director of Product Management discussed advanced topics including pain points with Amazon Lex, cost optimization strategies for Amazon Kendra, and vector databases. My understanding of these complex technologies was minimal. The emergence of generative artificial intelligence (AI) further heightened my insecurity. I was determined to enhance my role and contribute to solving complex problems through data-driven solutions, leading me to strengthen my ML skills through on-the-job learning.

My ML skilling journey

To begin, I adopted a self-directed learning approach, utilizing online courses, tutorials, and hands-on projects to bridge the gap in my knowledge. AWS Skill Builder has more than 600 free digital courses across the range of AWS service areas, for beginners to advanced learners, as well as subscription-only hands-on learning resources like AWS Cloud Quest or AWS Jam.

You don’t need a technical background or to have gone through intensive programs to start learning about machine learning. It’s an area that is beneficial for people across many roles to upskill in. I started with a few short ML fundamentals courses, each of which is beginner friendly.

After getting a foundational understanding of machine learning, I was ready to deepen my skills. I enrolled in the following AWS training courses which walked through key aspects of the machine learning pipeline and real-world applications:

Learning ML by doing

The more I learned about machine learning, both through courses and working with customers on real-world industry use cases, the more my curiosity grew. I started to think critically about the business problems ML could help solve and how I could apply my new skills to make an impact. This motivated me to supplement my learning with some hands-on mini-projects using sample datasets, such as building a simple classification model in SageMaker to predict customer churn. Though basic, these initial attempts at applied ML strengthened my comprehension and confidence. Learning by doing and stepping outside my comfort zone accelerated my skills and showed me how much potential there is to transform organizations with ML.

Preparing for AWS Certified Machine Learning- Specialty

The AWS Certified Machine Learning – Specialty exam is one of the hardest AWS certifications but I wanted to validate my newfound skills by passing the exam and earning this coveted industry credential. You may recall that I had previously earned the foundational certification, AWS Certified Cloud Practitioner. I recommend everyone start with this certification, and perhaps also earn an Associate level AWS Certification, before moving to the Specialty exams.

AWS does not require that you take specific prep ahead of the exam. However, I recommend that you follow these steps to prepare for exam day:

  • Review the exam guides to understand the exam content and take AWS Certification Official Practice Question Set, found in AWS Skill Builder, to understand exam style questions.
  • Learn about exam topics by enrolling in digital training on AWS Skill Builder.
  • Prepare for the exam by subscribing to AWS Skill Builder to access the self-paced Exam Prep Course (with Practice Materials). Review white papers and AWS services-related FAQs available on the exam page.
  • Verify your exam readiness by taking the AWS Certification Official Practice Exam available on AWS Skill Builder with a subscription.

Specific courses and resources I used included:

  • Machine Learning Specialization covers a comprehensive range of topics, including linear regression, machine learning algorithms, neural networks, deep learning, sequence models, and practical applications, providing a holistic understanding of the field. I went through this twice, re-doing all the hands-on labs to solidify my applied knowledge.
  • Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow is a book by Aurélien Géron and it boosted my understanding of several training models, including support vector machines, decision trees, random forests, and ensemble methods.
  • AWS Certified Machine Learning Specialty equips you with in-depth knowledge of ML on AWS through video lectures, hands-on labs, and practice exams to fully prepare you for the specialty exam and give you applied skills for building, training, and deploying ML models.
  • Machine Learning Plan – AWS Skill Builder provides interactive tutorials, videos, and labs to give you hands-on machine learning experience through developing models, training algorithms, and deploying projects on AWS to help you gain proficiency in machine learning on the cloud.
  • Amazon SageMaker Technical Deep Dive Series- AWS on YouTube offers detailed video tutorials and demonstrations by AWS machine learning experts to provide an in-depth look at SageMaker capabilities and teach you how to build, train, tune, deploy and manage machine learning models with.
  • Amazon SageMaker Developer Guide provides comprehensive documentation, sample code, and step-by-step tutorials to teach developers the complete capabilities of SageMaker for every step of the machine learning workflow from data prep and model training to deployment and monitoring.
  • AWS Machine Learning: Exam Preparation provides focused video tutorials, practice exams, and knowledge checks to fully prepare students for the AWS Certified Machine Learning Specialty exam by covering key ML concepts and AWS services required for the exam in an efficient and structured learning path.

On exam day, I felt anxious and even worried I might fail given the long, challenging format. Waiting for my results was nerve-wracking. But the next day, I found out I passed! After investing so much time and effort self-studying for around seven months, passing the exam gave me an immense sense of pride and achievement. It was incredibly rewarding to see my dedication result in successfully building this new body of machine learning knowledge. I’m motivated to continue expanding on this strong ML foundation.

Generative AI learning

Generative AI, a sub-field of Deep Learning, holds the capability to create new content autonomously. Unlike traditional models that rely on predefined patterns and rules, generative AI can autonomously generate human-like text, images, and even code. Exploring the nuances of generative models with the free digital course, Generative AI with Large Language Models and Generative AI Foundations on AWS, became an exciting part of my learning journey, illustrating how staying attuned to evolving technologies enriches one’s understanding and proficiency.

Conclusion

As ML continues to evolve the IT landscape, embedding into various industries and sectors, it becomes increasingly integral for driving innovation, enhancing efficiency, and unlocking new possibilities for data-driven decision-making. My commitment to continuous learning is unwavering. I eagerly soak up new knowledge and technologies to become proficient in areas that directly impact my role. My immediate plan is to explore advanced ML concepts and specialize in Responsible AI.

My journey in ML, starting from zero, has been a testament to the power of self-directed learning. If you’re embarking on a similar journey, my advice is to start with a growth mindset, remembering that every challenge is a stepping stone to propel you forward. The world of ML is vast and welcoming to those with diverse backgrounds, so make the leap today. Happy learning!

Additional blog and learning resources for ML, generative AI, and certification prep: