After passing the AZ-303/304 exams to earn the Microsoft Certified: Azure Solution Architect Expert certification, I decided it was time to take a slight detour to have a look at the other Microsoft Fundamentals Certifications. My interest in Artificial Intelligence (AI) took me down the path towards the Microsoft Certified: Azure AI Fundamentals certification.
Disclaimer: Microsoft reviews certifications regularly to help ensure that they remain relevant, technically accurate and that they assess the skills needed to thrive in a cloud-based world. Exams evolve over time and may change without notice. This post represents my best knowledge of the AI-900 exam as of Q1 2021, but you should always consult Microsoft’s Official AI-900 Exam Page for the contents of the exam.
What is the Azure AI Fundamentals Certification?
The Azure AI Fundamentals Exam is an opportunity to demonstrate knowledge of common Machine Learning (ML) and Artificial Intelligence (AI) workloads and how to implement them on Azure.
If you plan to take this exam you should have foundational knowledge of ML & AI concepts and related Azure services. The exam is intended for anyone (technical and non-technical) looking to begin working with ML and AI on Azure.
Azure AI Fundamentals can also be used to prepare for other Azure role-based certifications, but it is not a prerequisite for any of them.
How Did I Prepare?
I used my limited Azure ML & AI exposure from the past six years. I’ve only had limited involvement with projects using ML and AI services. This was a major driver in my choice to pursue the certification as I wanted to expand my knowledge so I could be better prepared to support future projects and opportunities. I spent about a week studying an hour or two a night for the exam, which worked out well.
What Resources Did I Use to Prepare?
I used the following resources to get ready:
- Microsoft Learn – Get started with artificial intelligence on Azure
- Microsoft Learn – Use visual tools to create machine learning models with Azure Machine Learning
- Microsoft Learn – Explore computer vision in Microsoft Azure
- Microsoft Learn – Explore natural language processing
- Microsoft Learn – Explore conversational AI
Unlike past use of MS Learn resources, I completed all the modules above due to my limited knowledge on the topics.
My Challenge Area
While I cannot talk about exam specifics for obvious reasons, I do want to point out the area that I struggled with the most:
Azure Machine Learning Designer related questions were the most challenging for me on this exam. I’m not sure why, but I find the concepts around models and pipelines difficult to fully understand. I’m looking forward to opportunities for real world projects that can help to flush out any confusion I still have on these topics.
This was my fifth exam taken remotely through PearsonVue due to COVID-19. I gave myself a week to prepare and study when I scheduled the exam continuing to follow my pattern of giving myself a deadline to force the work needed to pass.
Same procedure as the previous four experiences. I took the test in my home office. The night before I made sure to disconnect and clear out anything that might be an issue for the online proctor. I set up my laptop on my desk, provided pictures of my setup, driver’s license, and face, then sat in front of my computer while the proctor verified that everything was good to go. I never had to talk with them at all, the exam launched, and I cruised through it in around a half hour (you have a full hour for the test).
Immediately after the test, I received my detailed score breakdown in various areas and was told that I passed. An e-mail shows up no more than an hour later with details on getting my certificate, claiming it on Acclaim, sharing it on LinkedIn, etc.
Azure AI Fundamentals contains a surprising range of materials and requires a bit of depth in every area.
Here are a handful of other exams to consider if you’ve passed AI-900:
- AI-102: Designing and Implementing a Microsoft Azure AI Solution
- DP-100: Designing and Implementing a Data Science Solution on Azure
As for me, I moved on to DP-900 to help expand my data concept skills.