Agenticum · Thoughts · 7 min read

AI Engineering Courses for the UAE and Saudi Arabia: IBM vs Microsoft AI-103 vs AWS MLA (Late 2026)

Microsoft has retired AI-102, and its current AI engineering exam is AI-103. AWS has ended the English version of its MLA-C01 machine learning engineer exam and opened MLA-C02 in beta. Here is how three AI engineering paths compare in October 2026 for professionals and teams in the UAE and Saudi Arabia.

Nikolay Zuykov · Agenticum · 30 October 2026

What an AI engineering course covers in 2026

These three options cover different aspects of AI engineering. The IBM and Microsoft options both cover retrieval-augmented generation (RAG) and AI agents. IBM's certificate includes a course on building AI agents with RAG and LangChain. Microsoft's AI-103 exam tests RAG inside an application and orchestrated multi-agent solutions. The third option, from AWS, checks whether you can put machine learning workloads into production and keep them running.

They also lead to different types of credential. IBM's Coursera certificate is a structured course series. Microsoft AI-103 is a cloud-vendor exam for AI apps and agents on Azure. AWS MLA is a cloud-vendor exam for ML in production on AWS. To choose well, first decide which of these you need.

Option 1: IBM AI Engineering Professional Certificate (Coursera)

The IBM AI Engineering Professional Certificate is a 13-course series on Coursera. Coursera estimates it takes about 4 months at 10 hours a week. One course teaches you to build AI agents with RAG and LangChain, and Coursera says it takes 8 hours.

The series includes a capstone. You build your own generative AI application that you can show in job interviews. In the capstone project course, learners build a question-answering bot with LangChain and a large language model.

The IBM certificate programme is included in Coursera Plus. At US list price, Coursera Plus costs $59 a month (AED 216.68 / SAR 221.25) or $399 a year (AED 1,465.33 / SAR 1,496.25). The annual plan has a 14-day money-back guarantee. If you expect to finish within about four months, four monthly payments cost less than one annual plan.

Option 2: Microsoft Azure AI Apps and Agents Developer Associate (exam AI-103)

Microsoft has retired the Azure AI Engineer Associate certification (exam AI-102) and its renewal assessment. Its current AI engineering credential is the Azure AI Apps and Agents Developer Associate. This certification shows that you can design, build and deploy Azure AI solutions with Python and Microsoft Foundry. The exam is AI-103, Developing AI Apps and Agents on Azure.

Microsoft expects candidates to have experience building apps in Python. You should also know what general AI, generative AI and Azure services can do. The skills outline in force dates from 16 April 2026. It includes adding RAG to an application and building orchestrated multi-agent solutions.

The exam lasts 120 minutes, and you need a score of 700 or higher to pass. If you fail, you can retake it 24 hours after your first attempt. Microsoft does not publish one global price. The price depends on the country or region where the exam is proctored. Microsoft associate certifications expire every year, and you renew them by passing a free online assessment on Microsoft Learn. AI-103 is offered in English, Chinese (Simplified), Chinese (Traditional), French, German, Japanese, Korean, Italian, Portuguese (Brazil) and Spanish. Arabic is not one of them.

Option 3: AWS Certified Machine Learning Engineer - Associate (MLA-C01 / MLA-C02)

The last day to take the English version of MLA-C01 was 28 September 2026. Registration for the English-only MLA-C02 beta opened on 1 September 2026. Beta exams began on 29 September 2026.

The MLA-C02 beta costs $75 (AED 275.44 / SAR 281.25). The standard price listed for MLA-C01 is $150 (AED 550.88 / SAR 562.50). AWS charges less for beta exams than for standard exams and says taxes may be added. MLA-C02 has 85 multiple-choice and multiple-response questions. MLA-C01 has 65 multiple-choice and multiple-response questions. You can sit the exam at a Pearson VUE test centre or online with a remote proctor.

AWS says the ideal candidate has at least one year of experience in machine learning engineering or a related field, plus one year of hands-on AWS experience. The exam checks whether you can put ML workloads into production and run them.

Side-by-side comparison: cost in USD, AED and SAR, time, exam, focus

	IBM AI Engineering (Coursera)	Microsoft AI-103	AWS MLA-C02 (beta)
Focus	Course series with a generative AI capstone (RAG, LangChain)	AI apps and agents on Azure with Python and Microsoft Foundry	ML workloads in production on AWS
Price (USD)	$59 a month or $399 a year (Coursera Plus, US list price)	Set by the country where you sit the exam	$75 beta ($150 standard price listed for MLA-C01), tax may be added
Price (AED)	AED 216.68 a month or AED 1,465.33 a year	Set by country	AED 275.44 (AED 550.88 standard)
Price (SAR)	SAR 221.25 a month or SAR 1,496.25 a year	Set by country	SAR 281.25 (SAR 562.50 standard)
Time	About 4 months at 10 hours a week	120-minute exam	Check the AWS page
Format	13-course series, including a capstone project course	Proctored exam; passing score: 700; first retake available after 24 hours	85 questions; Pearson VUE test centre or online proctored
Recommended experience	Check the course page	Python app development; familiarity with general AI, generative AI and Azure services	1 year of ML engineering plus 1 year of hands-on AWS
Languages	Check the course page	10 languages, no Arabic	Beta in English only
Renewal	Check the course page	Every year, free online assessment	Check the AWS page

Currency note: USD is converted to AED at the peg of 3.6725 and to SAR at the peg of 3.75.

How to choose: a 4-step decision path

Which cloud does your employer or client run on? Consider AI-103 if your work involves building AI apps or agents on Azure. Consider the AWS MLA exam if your work involves deploying and operating ML workloads on AWS.
Do you need structured learning or a proctored credential? If you are new to the topic, a course series with a portfolio project gives you something to show. If you already do the work and need proof, a vendor exam is the more direct route.
Check your Python and cloud experience against each option. AI-103 expects experience building apps in Python. AWS recommends a year of ML engineering and a year of hands-on AWS. If you do not yet have the recommended experience for either exam, check the IBM course prerequisites on the course page before enrolling.
Fit the choice to your budget and time. Coursera estimates that the IBM certificate takes about 4 months at 10 hours a week. The MLA-C02 beta costs $75, and AI-103 is priced by country.

What to check before you pay

The live exam page. Microsoft has retired AI-102, and its current credential uses exam AI-103. The English version of MLA-C01 ended on 28 September 2026, and MLA-C02 is in beta.
The exam language. AI-103 is not offered in Arabic, and the MLA-C02 beta is in English only.
The local price. Microsoft prices AI-103 by country, AWS may add tax, and the Coursera prices above are US list prices.
Vouchers and licences. Ask your employer or its cloud partner whether it holds exam vouchers or Coursera seats before you pay yourself.
Renewal and retakes. Plan time for Microsoft's yearly renewal assessment. Read each provider's retake policy before your exam date.

For companies: training an AI engineering team

For a team, consider combining a course series with relevant vendor exams. One possible approach is to start with the IBM certificate or its RAG and agents course. Each engineer then takes the vendor exam that matches the company's cloud: AI-103 for Azure work, MLA for AWS work.

To assess how each engineer applies the training to your company's data, ask them to complete a practical project. For example, after the training each engineer could build a small internal question-answering prototype on a non-sensitive set of company documents, on the cloud you already run.