Back to Courses

Skills you leave with

What you'll learn

Build modern frontends with HTML, CSS, JavaScript, React, and Next.js
Design and ship a real backend: Node.js, Express, PostgreSQL, and Prisma
Integrate LLMs into your apps with structured outputs and streaming
Build RAG pipelines over your own documents with pgvector
Create AI agents with tool calling, memory, and MCP
Master agentic coding with Claude Code and Cursor
Evaluate, deploy, and run AI products in production
Launch your own capstone, live, with a public demo day

Prerequisites

A laptop and basic computer skills. That's it. No coding experience needed, we start from the very beginning.

Curriculum

Every section, every lesson

16 SECTIONS · 163 LESSONS

Set up your editor and build a clear mental map of how software and the web work, before writing any code.

  • 01What you will build in this course
  • 02What software is
  • 03How the web works
  • 04Frontend, backend, and database
  • 05What happens when you open a website
  • 06Installing Cursor, our editor
  • 07Terminal essentials

Meet your mentor

Your instructor

Abdule Yousuf

Abdule Yousuf

AI Researcher · Co-founder, Cassivelle

Full profile

Abdule is a senior software engineer with over six years of experience in software development who switched into AI research, now pursuing his M.Sc. in Artificial Intelligence at Universiti Malaya. The last two years of his work have been AI agents specifically, and they still are: at Cassivelle, the company he co-founded, he builds and ships agents daily, custom models and specialized agents deployed into parliaments, ministries, classrooms, and hospitals across the Horn of Africa.

He leads Ardaykaab Academy, which has trained over 3,000 students across Somaliland. He teaches the way he works: build real things, and be able to explain every line.

6+

years in software development

2+

years building AI agents

3,000+

students trained

The full story

About this course

Fullstack AI Engineer is a hands-on, project-driven program that takes you from zero to shipping real AI products. Instead of drowning you in theory, it builds one skill on top of another: first the web itself, then the AI layer, then your own launched product.

The early modules are strictly coding by hand: HTML, CSS, and JavaScript, then React and Next.js, then a real Node.js backend with Express, PostgreSQL, and Prisma. No AI-written code here, AI may explain a concept or an error, never write for you. That's deliberate: you build the foundational logic yourself so that later you can direct AI agents with confidence instead of depending on them.

Then the AI layer begins: LLM APIs, RAG pipelines on pgvector, AI agents with tool calling, evaluation, and agentic coding with Claude Code and Cursor. We build one complete AI agent together, step by step, from an empty folder to a working system.

You finish by building and launching your own capstone, deployed, evaluated, and live, and leave as an engineer who can build and ship real AI agents, with a GitHub portfolio to prove it.