I recently passed this exam after around two weeks dedicating the evenings to its preparation, along a big part of the weekends (if not all of them). I was already familiar with AI and the Machine Learning topic, ML.NET and had “played” with some of the services. At the moment I passed the exam, I could have done the AI-100 but wanted to focus on the ‘latest and greatest’ which is now the standard exam, being the old one retired.
On this post I will share with you how did I prepare for it during three weekends and two in-between weeks (only evenings). I will share not just a list of content or sites but the full content I picked, including some additions, along the strategy and how I made it work for me.
And a special thank-you to Florian…
I was the other day talking with a colleague from an earlier job that I had the luck to work with. He has recently finished his thesis and BSC on machine learning and we started talking about the Azure AI certification and that it could be interesting to detail how to prepare for this AI-102 exam… So, that was my motivation for this post :) - Florian This one’s for you :)
The AI-102 Exam
It is important to get familiar with the exam contents and the skills it measures, which from the certification site, states:
Microsoft Certified: Azure AI Engineer Associate Candidates for the Azure AI Engineer Associate certification should have subject matter expertise building, managing, and deploying AI solutions that leverage Azure Cognitive Services, Azure Cognitive Search, and Microsoft Bot Framework. Their responsibilities include participating in all phases of AI solutions development—from requirements definition and design to development, deployment, maintenance, performance tuning, and monitoring. Azure AI Engineers work with solution architects to translate their vision and with data scientists, data engineers, IoT specialists, and AI developers to build complete end-to-end AI solutions. Candidates for this certification should be proficient in C#, Python, or JavaScript and should be able to use REST-based APIs and SDKs to build computer vision, natural language processing, knowledge mining, and conversational AI solutions on Azure. They should also understand the components that make up the Azure AI portfolio and the available data storage options. Plus, candidates need to understand and be able to apply responsible AI principles.
In short, the skills measured:
- Plan and manage an Azure Cognitive Services solution (15-20%)
- Implement Computer Vision solutions (20-25%)
- Implement natural language processing solutions (20-25%)
- Implement knowledge mining solutions (15-20%)
- Implement conversational AI solutions (15-20%)
So, what’s the plan?
I wanted something I could get quickly a good overview of the concepts and get a general and in-detail feeling, and then, dig into the topics in depth!!
I wanted to go through that content deeply on the first weekend, then do some recaps on the second and go deep again on the third weekend.
Meanwhile I wanted fundamental hands on at the evenings and some reading and study, for around one hour or at most two with some pause in between.
So, summarizing:
- Full dive in on the first weekend with multimedia content.
- Mix of hands-on and reading during the week.
- Recap of the first week of training and refresh of topics during the weekend.
- Repeat the same strategy of the earlier week & weekend.
- And then, go for the exam!
Over the first weekend I did that and picked a video training which with I could familiarize more on the topic
Alrighty! and what content?
For the multimedia content, I had two contenders:
- The AI-102 Course fro Scott Duffy, at Udemy (6h of content)
- The AI-100 Certification path, at Pluralsight (36h of content)
Just note that Scott’s course had some downloadable materials and reference to other content that proved to be very valuable.
I decided to go for Scott’s course mainly and use Pluralsight to reinforce some of the points that might not be clear.
As an added tip for the video, I always reproduce it between 1.5x and 2x speed so you cannot put your focus in anything else or you get lost… with the benefit of consuming the content faster.
I also disable computer and mobile notifications so to have full focus and try to time the sessions with a pomodoro clock.
For the hands-on I went with the Microsoft Learn recommendation but I added some of their content on top.
My customized picks were, in order:
- Introduction
- Concrete services
- Evaluate text with Azure Cognitive Language Services (2 h 13 min)
- Process and classify images with the Azure cognitive vision services (3 h 15 min)
- Explore computer vision in Microsoft Azure (2 h 56 min)
- Create computer vision solutions with Azure Cognitive Services (4 h 27 min)
- Explore natural language processing (2 h 13 min)
- Process and translate text with Azure Cognitive Services (1 h 48 min)
- Process and Translate Speech with Azure Cognitive Speech Services (1 h 45 min)
- Process natural language with Azure Cognitive Language Services (2 h 51 min)
- Create conversational AI solutions (2 h 24 min)
- Build a bot with QnA Maker and Azure Bot Service (29 min)
- Create a Language Understanding solution (2 h 28 min)
- Implement knowledge mining with Azure Cognitive Search (2 h 35 min)
- Extract text from images and documents (1 h 31 min)
Note that some of the “learning paths” linked here, might contain overlapped, shared modules.
On another topic, if a module/section contains a link to the product page, it is a good idea to go there, and study it, read some sections and become familiar with it. And maybe to peek at some quickstarts, annotate them and once done the recommended MS-Learn trainings AND LABS, come back to these links and do the most relevant quickstarts. You learn by doing, that’s what makes us, humans, remember.
And Good luck!!
I did what is stated on this Certification preparation guide and got a nice score, so if I did… so can you ;)
Hope you enjoyed the Certification preparation guide, and If you believe I missed any tip or it can be improved, please let me know so I can improve it!
Have fun!