Agenticum · Thoughts · 6 min read

Data Science Certifications in the UAE and Saudi Arabia (Late 2026): Prices in USD, AED and SAR

Google Cloud, AWS and Databricks each offer a machine learning certification, and Coursera hosts a beginner IBM course series. Microsoft has retired its Azure Data Scientist Associate certification, and AWS held the last exam day for its Machine Learning - Specialty on March 31, 2026. Compare the exam fees, formats and experience recommendations below; check Microsoft and Coursera for their current prices.

Nikolay Zuykov · Agenticum · 1 November 2026

Short answer: three vendor certifications and one beginner route

If you already work with data and want a platform-specific exam, start with the three certifications in the table below; Microsoft’s Azure credential is covered separately. If you are starting from zero, begin with a course series instead.

Certification	Fee (USD)	Exam	Valid for	Recommended background
Google Cloud Professional Machine Learning Engineer	$200	2 hours, 50-60 questions	2 years	3+ years of industry experience, including 1+ year on Google Cloud
AWS Certified Machine Learning Engineer - Associate (MLA-C02, beta)	$75 (beta price)	170 minutes, 85 questions	3 years	At least 1 year using Amazon SageMaker AI and Amazon Bedrock
Databricks Certified Machine Learning Associate	$200	90 minutes, 48 scored questions	2 years	No prerequisites; Databricks strongly recommends related training

For beginners, the IBM Data Science Professional Certificate on Coursera is a series of 12 courses that requires no prior experience. Coursera estimates about 4 months at 10 hours a week.

Side-by-side comparison: price, format, validity, languages

Exam	USD	AED	SAR	Minutes	Questions	Validity	Languages
Google Cloud Professional ML Engineer	200	734.50	750.00	120	50-60	2 years	English, Japanese
AWS MLA-C02 (beta)	75	275.44	281.25	170	85	3 years	English only during the beta
AWS MLA-C01	150	550.88	562.50	130	65	3 years	Korean, Japanese, Simplified Chinese (English ended)
Databricks ML Associate	200	734.50	750.00	90	48 scored	2 years	English, Japanese, Brazilian Portuguese, Korean

You can take the Google exam remotely with online proctoring or at a testing center. The Databricks exam is delivered online or at a test center.

The exam for Microsoft's Machine Learning Operations Engineer Associate credential, AI-300, lasts 120 minutes and is offered in English only. Its price depends on the country where the exam is proctored.

Note: AED and SAR figures use the fixed pegs of 3.6725 AED and 3.75 SAR to 1 USD. Google lists its fee as "plus tax where applicable", so check the final amount at checkout. Arabic is not among the listed languages for any of these exams.

Recent changes

Microsoft Certified: Azure Data Scientist Associate (DP-100) and its renewal assessment are retired. Microsoft's current ML credential on Azure is Machine Learning Operations Engineer Associate (exam AI-300). It covers MLOps and GenAIOps on Azure Machine Learning and Microsoft Foundry.
AWS Certified Machine Learning - Specialty had its last exam day on March 31, 2026. If you already hold it, you stay certified for 3 years from the date you earned it, but you can no longer book it.
AWS MLA-C01 in English ended on September 28, 2026. English-speaking candidates now take the MLA-C02 beta, which began on September 29, 2026 at $75. AWS has not yet set a general availability date. MLA-C01 stays available in Korean, Japanese and Simplified Chinese until then.
The Google exam now covers Gemini Enterprise Agent Platform instead of Vertex AI. It also reflects updates to Google Cloud's data and analytics stack and gives priority to Google Cloud native solutions.

Starting from zero vs. proving existing skills

Starting from zero. The IBM Data Science Professional Certificate on Coursera is a beginner-level series of 12 courses with no prior experience required. Coursera estimates about 4 months at 10 hours a week. Check the current price on Coursera before you plan a budget.

Proving existing skills. The vendors give the following preparation recommendations:

Google sets no formal prerequisites and recommends 3+ years of industry experience, including at least 1 year designing and managing solutions on Google Cloud.
AWS recommends at least 1 year of experience with Amazon SageMaker AI and Amazon Bedrock for MLA-C02.
Databricks sets no prerequisites but highly recommends related training.
Microsoft AI-300 expects a data science background with Python and entry-level DevOps skills, such as GitHub Actions and command-line interfaces.

Consider completing a beginner course series, practising on one platform, and then taking its certification exam.

The real cost: renewals and retakes

When budgeting, check whether renewal or retakes will involve additional fees.

Google: valid 2 years. You can renew through a Google Skills renewal path by completing designated courses or skill badges during the final year your certification is valid.
AWS: valid 3 years.
Databricks: valid 2 years. To stay certified you must take the current version of the exam again.
Microsoft: if you fail a certification exam, you can retake it 24 hours after the first attempt. Check the retake rules for the other vendors before booking.

Fee per year of validity (our calculation from the fees and validity periods above, first certification only):

Exam	USD per year	AED per year	SAR per year
Google Cloud Professional ML Engineer	100.00	367.25	375.00
AWS MLA-C02 (beta price)	25.00	91.81	93.75
AWS MLA-C01	50.00	183.63	187.50
Databricks ML Associate	100.00	367.25	375.00

Example: if the Databricks fee stays at $200, holding the certification for 4 years means two sittings, $400 in total (AED 1,469.00 or SAR 1,500.00).

How to choose: match the cloud your employer or target employer runs

Find out which platforms your target employers use. Read job ads from the UAE and KSA companies on your list and note whether they ask for AWS, Google Cloud, Databricks or Azure.
Check your experience against the vendor's recommendation. Google recommends 3+ years in industry; AWS recommends at least 1 year with SageMaker AI and Bedrock. If you have less experience than Google or AWS recommends, consider the beginner IBM course series. Databricks has no formal prerequisites but strongly recommends related training.
Check the exam language. The AWS MLA-C02 beta and Microsoft AI-300 are English only. Google offers English and Japanese; Databricks offers English, Japanese, Brazilian Portuguese and Korean.
Budget for the fee plus renewal. Use the AED and SAR figures above, add tax where it applies, and plan for renewal every 2 years (Google, Databricks) or 3 years (AWS).

For companies: certifying a data team

For employers in the UAE and Saudi Arabia, the same steps apply at team level. Certify people on the platform you actually run in production. Google, for example, describes its certified engineer as someone who builds, evaluates, productionizes and optimizes AI solutions using Google Cloud capabilities.

Alongside the platform certification, develop skills you can apply across different tools and cloud providers. Put renewal dates in the team calendar, since two of the three vendor certifications above lapse after 2 years.