What should you learn before an AI degree?
The key idea
Prepare three foundations together: programming, mathematics and careful reasoning about data. Using an AI application is different from studying how a model is built, evaluated and limited. A degree may require you to work with all three foundations.
Check your own university's syllabus in Egypt. The programming language, mathematics sequence and specialist modules differ between programmes. A C and C++ foundation is one possible preparation route, but it is not the entire AI curriculum.
Build a preparation map
| Foundation | Starting task | What to explain |
|---|---|---|
| Programming | Read a small table and count categories | Which rule determines each count |
| Algebra and functions | Plot a simple relationship by hand | Inputs, outputs and how one changes |
| Probability and statistics | Compare a count with its denominator | The population your percentage describes |
| Data habits | Label fictional examples consistently | What is missing or ambiguous |
Follow your programme's mathematics requirements, including linear algebra or calculus where applicable. Google's machine-learning preparation guidance also highlights mathematical and programming prerequisites; a specific tool's requirements should not be mistaken for a universal degree syllabus.
Why accuracy needs context
Imagine 100 fictional examples: 90 belong to category A and 10 to category B. A system that always predicts A is 90% accurate overall, yet it identifies none of the B examples.
The arithmetic is straightforward, but the lesson is important: a headline percentage does not explain every kind of mistake. Ask which examples were tested, what each category means, and which errors matter for the intended task.
Try a second system that correctly identifies 85 A examples and 7 B examples. Its overall accuracy is 92%, and it finds 7 out of 10 B examples. These measures answer different questions.
Use programming practice to understand the calculation
Represent the correct and incorrect counts with variables. Write a short explanation of how the totals relate. Check that all counts add up to the stated sample size before calculating percentages.
Use invented data for your first exercises. Do not collect sensitive personal data simply to make a beginner exercise look realistic. The purpose is to understand the calculation and its assumptions.
Where the Gammal Tech courses fit
C and C++ support general programming foundations. Data Structures and Algorithms extend your ability to organise data and reason about computational work. OOP develops a way to organise state and behaviour. They are foundations you can use alongside the mathematics and other languages required by your institution.
Before the first semester, keep one small data exercise, one programming exercise and a list of mathematics topics to review. Your goal is to arrive ready to learn and question results carefully, without treating course completion as a guarantee of university grades or AI expertise.
References and course details
Use these references for further study. The examples and exercises in this article are educational; consult the catalogue for course details.