Agenticum · Thoughts · 6 min read

Machine learning courses for beginners in the UAE and Saudi Arabia: three options compared for 2026

This guide compares three beginner machine learning courses: the Machine Learning Specialization from DeepLearning.AI and Stanford Online, Google's Machine Learning Crash Course and Microsoft's ML-For-Beginners. We set them side by side on the same criteria, give prices in AED and SAR, and suggest which one fits your profile.

Nikolay Zuykov · Agenticum · 4 November 2026

Three beginner machine learning courses, side by side

The table compares the three courses using the same criteria. Prices are in USD, with AED and SAR amounts at the fixed pegs (see the price box below). The sections after the table explain each row.

Course	Provider	Hours	Level	Tools	Price (USD / AED / SAR)	Certificate	Arabic support
Machine Learning Specialization	DeepLearning.AI and Stanford Online	About 95, across 3 courses	Beginner: basic coding and high-school maths	Python with NumPy, scikit-learn and TensorFlow	49 / 179.95 / 183.75 a month on Coursera, or DeepLearning.AI Pro from 25 / 91.81 / 93.75 a month	Through a Coursera subscription or DeepLearning.AI Pro; none if you audit for free on DeepLearning.AI	Not confirmed. Taught in English, with subtitles in 30 languages
Machine Learning Crash Course	Google	About 15, across 12 modules	Beginner: basic algebra and statistics, Python helps	Python exercises in Colab, in the browser	Free, according to Google's relaunch announcement	None stated in our sources	Not stated in our sources
ML-For-Beginners	Microsoft Cloud Advocates	12 weeks, 26 lessons	Beginner, no deep learning	Mainly scikit-learn	Not stated in our sources	None stated in our sources	Arabic translation, one of 50+ languages

Option 1: the Machine Learning Specialization (DeepLearning.AI and Stanford Online)

DeepLearning.AI and Stanford Online built this Specialization together. Andrew Ng created it with Eddy Shyu, Aarti Bagul and Geoff Ladwig. DeepLearning.AI last updated its course page on 2 September 2026.

The Specialization has three courses. Coursera lists them at 33, 34 and 28 hours, about 95 hours in all, and estimates about 2 months at 10 hours a week. You build and train models in Python using NumPy, scikit-learn and TensorFlow. Coursera says you should understand basic coding (for loops, functions, if/else statements) and high-school maths. The teaching is in English, with subtitles in 30 languages. We could not confirm that Arabic is one of them.

There are two ways to pay. On Coursera you pay a subscription of USD 49 a month for this Specialization. Coursera Plus, the all-access plan, costs USD 59 a month after a 7-day free trial. Coursera also offers financial aid to learners who cannot afford the fee. The other route is a DeepLearning.AI Pro membership: USD 25 a month billed annually or USD 30 a month billed monthly. DeepLearning.AI states that a free audit does not include a certificate, so you need one of the paid routes if you want the credential.

Example: if you keep to Coursera's estimate of two months, the subscription comes to USD 98, which is AED 359.91 or SAR 367.50 at the pegs.

Option 2: Google's Machine Learning Crash Course

Google calls this course its "fast-paced, practical introduction to machine learning". It includes animated videos, interactive visualisations and hands-on exercises. When Google relaunched it, the announcement called it a free, online 15-hour self-study course and said the new version adds large language models and AutoML. Check the course page before you plan around the price.

The course has 12 modules, and Google recommends that newcomers take them in order. The modules include Neural Networks, Embeddings, Intro to Large Language Models, AutoML and ML Fairness. Programming exercises run in the browser on Google's Colaboratory platform, with no setup.

Google expects you to be comfortable with variables, linear equations, graphs of functions, histograms and statistical means. Some Python experience helps because the exercises are in Python, but Google does not make it a strict requirement. This makes the course a good first step if you want the concepts quickly before you pay for anything.

Option 3: Microsoft's ML-For-Beginners curriculum

Cloud Advocates at Microsoft publish ML-For-Beginners as a 12-week, 26-lesson curriculum about machine learning. It mainly uses scikit-learn and leaves out deep learning.

The repository includes more than 50 translations, Arabic among them. Microsoft notes that the translations make the download much bigger. For a learner who prefers to read in Arabic, this is the only one of the three courses where our sources confirm an Arabic version. For a workplace study group, use the 12-week curriculum as a starting point and adjust the schedule to your group.

What it costs in AED and SAR

AED and SAR amounts are at the fixed pegs to the US dollar.
Item	USD	AED	SAR
Coursera subscription for the Specialization, per month	49	179.95	183.75
Coursera Plus, per month after a 7-day trial	59	216.68	221.25
Coursera Plus Annual, offer price (usual price)	239.40 (399)	879.20 (1,465.33)	897.75 (1,496.25)
DeepLearning.AI Pro, per month billed annually	25	91.81	93.75
DeepLearning.AI Pro, per month billed monthly	30	110.18	112.50
AWS MLA-C02 exam, beta price	75	275.44	281.25
AWS MLA-C01 exam, previous version	150	550.88	562.50

The AED and SAR figures use the fixed pegs, not live market rates, and are converted from USD prices. The amount you see at checkout from the UAE or Saudi Arabia may differ, and DeepLearning.AI says taxes may apply depending on your location.

Which machine learning class fits you: a decision guide

You are a curious professional or a manager who wants the concepts. Start with Google's Machine Learning Crash Course. It takes about 15 hours and covers large language models and AutoML, so you can judge whether you want to go further before you spend money.
You are changing careers and want a certificate. Take the Machine Learning Specialization through Coursera or DeepLearning.AI Pro. Plan for about 95 hours. If the fee is a barrier, apply for Coursera's financial aid.
You prefer to learn in Arabic, or you run a study group at work. Use Microsoft's ML-For-Beginners. The Arabic translation helps, and you can use the 12-week, 26-lesson plan as a starting point for your group's schedule. Add a deep learning course later, because this curriculum leaves it out.
You are an engineer already working on AWS. Consider the AWS exam described below once you have the hands-on experience AWS recommends.

An AWS exam and a Saudi programme to know about

If you have the AWS experience described below, consider the AWS Certified Machine Learning Engineer - Associate exam. Registration is open for the updated exam, MLA-C02, in beta. During the beta it is offered in English only and costs USD 75 (AED 275.44 / SAR 281.25). AWS labels this beta pricing, so check the price again before you book. The exam has 85 multiple choice and multiple response questions in 170 minutes. You can take it at a Pearson VUE testing centre or online with a proctor, and the certification is valid for 3 years. The previous version, MLA-C01, cost USD 150, and 28 September 2026 was the last day to take it in English. AWS recommends at least a year of experience with Amazon SageMaker AI, Amazon Bedrock and other AWS machine learning services before you sit MLA-C02.

In Saudi Arabia, SDAIA Academy is an initiative of the Saudi Data and AI Authority. According to Saudipedia, registration is free for men and women aged 18 to 60. We could not confirm current machine learning cohorts, dates or eligibility, so check SDAIA's official channels before you plan around a programme.