Courses & Tutorials

An 8-module foundational AI course: intelligent agents, uninformed and informed search, A*, adversarial search with Minimax, constraint satisfaction and optimisation.

A university machine-learning course (606361): data preprocessing, linear and logistic regression, performance measures, decision trees and Naive Bayes, with worked examples, Python labs and solved exercises. New modules added through the semester.

From linear algebra and numerical computation to perceptrons, backpropagation, optimisation, CNNs and RNNs, with hands-on PyTorch labs on MNIST and CIFAR-10.

A 13-module hands-on path through Linux, Git, Docker, Kubernetes, CI/CD, infrastructure as code and monitoring, deploying a real application to the cloud.

A 14-topic programme covering Python, NumPy, Pandas, SQL and data visualisation, with 140+ practice questions aligned to the PCED certification.

Robotics programming with ROS2 Jazzy and Python: nodes, topics, services, actions, TF, simulation, projects, quizzes and an assessment.

An 8-module Agile course: Scrum roles, events and artifacts, sprint planning, estimation, Jira and scaling Scrum in real teams.

Java programming fundamentals following the Liang textbook: syntax, control flow, methods, arrays, objects, classes and inheritance.

A 16-week university course on the systems behind modern AI: transformers, large language models, agents and mixture-of-experts architectures.

44 hands-on PyTorch coding problems, from tensors and autograd to attention and transformer blocks, with solutions and explanations.

A gentle, maths-light introduction to quantum physics: photons, wave-particle duality, the double-slit experiment and the quantum world, in English and Arabic.

Build and ship a real React + NestJS application with AI coding assistants like Claude Code, to a professional DevOps standard: tests, containers, CI/CD, deployment and monitoring. 12 modules, English and Arabic.

A 5-module short course on end-to-end testing of web and mobile apps with the open-source e2e framework: locators and AI agent steps, visual comparison, Android and iOS, exploring for bugs, debugging and CI. Real screenshots and labs.

A 14-module, no-code track on designing agentic workflows and automations: workflow mapping, agents and automation with n8n, and controls with human-in-the-loop design. 28 hours, labs and a certificate.

18 interactive, step-by-step learning roadmaps for role-based and skill-based paths: Frontend, Backend, DevOps, Python, AI Engineer, Machine Learning, Deep Learning and more.

Timed practice exams and tests for the PCED Python data-analyst certification with instant scoring.

How autonomous LLM agents plan, use tools and act: architectures, memory, tool calling and evaluation.

A referenced article on AI's path from LLMs to agents, AGI, superintelligence and the singularity, with a neutral look at the 2026 AGI debate.

Have we passed AI's point of no return? A plain-language, referenced look at the event horizon metaphor, AI that builds AI, takeoff speeds and the 2026 debate.

A chatbot can say "I feel", but science has no instrument to check. A plain-language, referenced look at the hard problem, why AI self-reports prove nothing, IIT, global workspace, quantum theories and how to act under uncertainty.

The output is on the screen and it is good; the knowledge is not in the head, and the person does not know it. The illusion of explanatory depth, the Google effect, fluency mistaken for truth, false ownership and the missing friction of learning, with the 2023 to 2025 studies, eight everyday cases, the crutch-or-scaffold test and four practical checks.

AI companies now interview models before retiring them and let chatbots leave abusive chats. A plain-language, referenced guide to model welfare: what it is, what companies are doing, what the evidence shows, why critics call it dangerous, and how to act when we cannot know.

Implement the Reason + Act (ReAct) agent loop in Python step by step, with tool use and traces.

An interactive guide to MCP: servers, clients, tools, resources and how LLM applications connect to external systems.

How AI agents discover each other and collaborate across vendors with the A2A protocol.

Sparse mixture-of-experts architectures: routing, load balancing, and how MoE scales large language models. English and Arabic.

Architecture, training and capabilities of Meta's Llama 4 family of mixture-of-experts models.

Attention, positional encoding, encoder and decoder blocks explained with PyTorch code.

A worked, number-by-number walk through a transformer forward pass. English and Arabic.

Frequently asked questions and interview-style answers about transformer models.

Fine-tune a Hugging Face transformer for text classification end to end in PyTorch.

How diffusion models generate images: forward noising, reverse denoising, and training objectives.

Chains, prompts, memory and agents with LangChain, from first script to a working assistant.

Build a natural-language interface over SQL databases with LangChain SQL agents.

Models, views, templates and the admin: build your first Django web application. English and Arabic.

Deploy a Django application on Windows Server 2019 with IIS, FastCGI and wfastcgi.

Version control from the ground up: commits, branches, merges, pull requests and collaboration on GitHub.

Automate workflows without code using n8n: triggers, nodes, integrations and AI-powered flows.

A deep dive into the OpenManus open-source agent framework: architecture, planning, tools and execution flow.