Velamuvi AIAcademy
  • Getting started
  • Study plan

How to Start Learning AI From Scratch: A Realistic Plan

A clear, honest plan for complete beginners: what to learn first, what to skip for now and how to stay consistent.

· 3 min read · Velamuvi AI Academy

Learning AI looks overwhelming from the outside. There are hundreds of tools, new models every month and a lot of maths. The good news: you do not need all of it to begin, and the order you learn things in matters far more than how fast you go.

Step 1: Learn Python first

Almost every AI tool is used from Python, so it is the one skill everything else depends on. You do not need to become an expert. You need to be comfortable with variables, loops, functions, lists and dictionaries.

Start with Python for Beginners. Every example runs in your browser, so you can begin right now without installing anything.

Step 2: Decide what you want to build

"AI" covers several different jobs, and they need different skills.

  • If you want to build apps on top of models like Claude or GPT, you need APIs and prompt engineering before you need much maths. Follow the AI Developer roadmap.
  • If you want to train models yourself, you need statistics, linear algebra and machine learning. Follow the Machine Learning Engineer roadmap.
  • If you enjoy finding answers in data, the Data Scientist roadmap is the best fit.

Picking a direction early stops you from trying to learn everything at once.

Step 3: Learn the maths when you need it

Many beginners stop because they think they must finish a maths degree first. You do not. Learn enough to understand what a model is doing, then go deeper when a topic requires it. Our maths courses explain ideas with pictures and Python instead of proofs.

Step 4: Build small projects early

You remember what you use. After each course, build something small: a number guessing game, a spam detector, a chatbot. The project library has briefs with milestones for every level.

Step 5: Be consistent, not heroic

Thirty minutes a day beats five hours once a week. Short, regular sessions give your brain time to absorb new ideas.

What to skip for now

  • Choosing the "best" framework. The ideas transfer. Learn one well.
  • Buying a powerful computer. Early courses run in your browser, and free cloud notebooks cover deep learning.
  • Reading research papers. They will make sense later, after the fundamentals.

Where to go next

Open the first Python lesson and write one line of code. That is the whole first step.