ATHAR ACADEMY · Learn & Build16 Recorded Tracks + 7 Claude + Codex Live Sessions
Professional Applied AI Bootcamp

Build a Complete AI System.
With Claude + Codex.

16 self-paced tracks covering Python, AI, ML, LLMs, RAG, and agents, followed by 7 live sessions with Mostafa using Claude and Codex to plan, build, test, and deploy one connected WordPress site, web app, native mobile app, and AI backend.

ClaudePlan · Design · Review✕CodexBuild · Test · Refine
16Recorded tracks
07Live sessions
01Integrated system

LIVE TRAINING · CHOOSE YOUR START DATE

Choose Your Live Cohort with Mostafa

Access the self-paced recorded curriculum anytime. The seven Claude + Codex live-build sessions with Mostafa have two start-date options. Choose yours before checkout.

Selected live cohort Select 1 November or 1 December above

Your choice will appear in the cart and be saved in your WooCommerce order.

Select a cohort to continue

✦ These are cohort start dates, not the weekly session schedule. The private option opens a separate course and does not add the group bootcamp to the cart.

TWO LEARNING EXPERIENCES

Two Formats. One Connected Learning Journey.

The recorded track builds depth. The live track turns knowledge into a complete working project. Each serves a different purpose.

01 · RECORDED · SELF-PACED

From Foundations to AI Engineering

16 structured tracks covering Python, data, ML, deep learning, computer vision, transformers, LLMs, RAG, agents, MCP, and deployment.

  • Recorded lessons at your own pace
  • Practice tasks and topic projects
  • Optional reinforcement learning track
View 16 tracks ↗
02 · LIVE · WITH MOSTAFA

7 Sessions to Build & Deploy One System

Use Claude for product architecture, structured requirements, and critical review, then use Codex to build, test, and debug FastAPI, WordPress, web, and React Native components before integrating and deploying the MVP.

  • Instructor-led, hands-on implementation
  • A concrete deliverable after every session
  • GitHub project and demonstrable system
View 7 sessions ↗

INTERACTIVE SYSTEM BLUEPRINT

Explore the Architecture You’ll Build

Select any component to see its role in the connected AI system.

ATHAR / FULL STACK / AI● SYSTEM BLUEPRINT
SELECTED COMPONENT

AI Engine & APIs

A FastAPI backend connecting the LLM, RAG, and permitted tools with access checks.

Illustrative architecture for the live build project.

LIVE BUILD LAB · 7 SESSIONS

7 Live Sessions. Build with Claude + Codex.

Select a session to see the Claude planning task, Codex implementation task, hands-on prompts, and deliverables. Copy a prompt to adapt it for your project.

01 · ClaudeUnderstand the problem, write requirements, design architecture, and review decisions.
02 · CodexCreate files, implement features, test code, and resolve reproducible bugs.
03 · MostafaExplain decisions, review output, validate security, and approve a testable build.
LIVE SESSION 01 / 07

From an Idea to an AI System Architecture

DISCOVER · DESIGN

What will we build?

  • Hands-on Claude-assisted planning and review, plus Codex-assisted implementation and testing under instructor supervision.
  • Practical foundation: GenAI, LLMs, agents, RAG, and the difference between using a tool and building a product.
  • Select a real use case and scope the first MVP around users and workflows.
  • Design one connected WordPress + web app + native app + AI API architecture.
  • Use AI-assisted wireframing for pages, roles, and data flows.
  • Set up a GitHub repository, execution plan, and reviewed AI-assisted code workflow.
  • Establish privacy, licensing, and data-security boundaries.

How Claude and Codex Will Be Used

ClaudePlanning & Review

Turn the learner’s idea into an MVP specification: target users, stories, flows, system diagram, API contract, risks, and acceptance criteria.

Example prompt
Act as a solution architect. Specify an MVP with WordPress, a web app, React Native, FastAPI, and an AI service. Provide user flows, a component diagram, API contracts, access rules, and testable acceptance criteria. Ask about missing assumptions.
CodexDevelopment & Testing

Scaffold the GitHub monorepo for WordPress, web, mobile, and API, with README, starter configuration, and basic smoke tests.

Example prompt
Scaffold a monorepo with wordpress/, web/, mobile/, and api/. Add README, .env.example, starter run scripts, and a smoke test per component. Never commit real secrets. Show the planned file changes before large edits.

Human review checkpoint: Review feasibility, scope, and Codex-generated files before accepting them.

✓ SESSION DELIVERABLE

Product brief + wireframes + architecture diagram + GitHub repository.

↗ BETWEEN-SESSION PRACTICE

Define your use case and map the MVP screens and user journey.

RECORDED LEARNING · 16 TRACKS

Your Complete Recorded Curriculum

Search or filter all 16 tracks, and open any track to see its topics and practical activities.

01Python Programming Foundations

TOPICS

  • Variables and data types
  • Conditional statements and program flow control
  • Loops and iteration
  • Functions and code organization
  • Working with lists, dictionaries, and sets
  • Object-oriented programming (OOP) fundamentals
  • Error handling and exceptions
  • Modules and Python packages
  • Creating and managing virtual environments
  • Reading and writing files
  • Working with TXT, CSV, and JSON files
  • Text processing with regular expressions (Regex)
  • Git and GitHub fundamentals
  • Organizing software project files

PRACTICE & LABS

  • Build a simple student-record management application
  • Read and perform initial cleaning on a CSV file
  • Extract structured information from text with Regex
  • Publish a simple project to GitHub
02Data Management, SQL & APIs

TOPICS

  • Introduction to data management
  • Data types and data sources
  • Relational database fundamentals
  • Designing tables and relationships
  • SQL fundamentals
  • SELECT, WHERE, and ORDER BY queries
  • Working with different types of JOIN
  • GROUP BY and HAVING clauses
  • Aggregate functions
  • Creating and modifying database tables
  • Introduction to NoSQL databases
  • Differences between SQL and NoSQL
  • Working with Excel and CSV data
  • Consuming application programming interfaces (APIs)
  • Understanding JSON and REST APIs
  • Sending and receiving data using APIs
  • Web scraping techniques and responsible data collection
  • Building a simple data pipeline

PRACTICE & LABS

  • Design a simple e-commerce database
  • Write queries for sales reports
  • Consume an external API and convert results to a DataFrame
  • Collect permitted website data and perform an initial analysis
03Data Analysis & Applied Statistics

TOPICS

  • Introduction to data analysis
  • The lifecycle of a data analysis project
  • Descriptive statistics fundamentals
  • Mean, median, and mode
  • Variance and standard deviation
  • Statistical distributions
  • Introduction to probability
  • Correlation
  • Analyzing relationships between variables
  • Basic hypothesis testing
  • NumPy fundamentals
  • Working with arrays
  • Using Pandas to organize and analyze data
  • Data cleaning
  • Handling missing values
  • Detecting and working with outliers
  • Exploratory data analysis (EDA)
  • Creating charts with Matplotlib
  • Statistical visualizations with Seaborn
  • Interactive visualizations with Plotly
  • Association rule analysis
  • Discovering patterns in data

PRACTICE & LABS

  • Analyze student performance
  • Analyze an e-commerce dataset
  • Analyze customer behavior
  • Measure marketing campaign performance
  • Analyze car sales data
  • Analyze monthly sales and calculate KPIs
04Machine Learning

TOPICS

  • Introduction to machine learning
  • AI versus machine learning versus deep learning
  • Types of machine learning
  • Supervised learning
  • Unsupervised learning
  • Preparing data before model building
  • Splitting data into training and testing sets
  • Feature engineering
  • Encoding text and categorical variables
  • Scaling and normalization
  • Understanding data leakage
  • Regression models
  • Linear regression
  • Evaluating regression models
  • Classification concepts
  • Logistic regression
  • Naive Bayes
  • Decision trees
  • Random forests
  • K-nearest neighbors (KNN)
  • Support vector machines (SVM)
  • Clustering
  • K-means
  • Association rules
  • Recommendation systems fundamentals
  • Model performance measurement and evaluation
  • Accuracy, precision, recall, and F1 score
  • Confusion matrix
  • Cross-validation
  • Overfitting and underfitting
  • Building pipelines with Scikit-learn

PRACTICE & LABS

  • Predict house prices
  • Predict customer churn
  • Classify spam emails
  • Predict response to banking offers
  • Segment customers
  • Build a basic product recommendation engine
05End-to-End ML Projects

TOPICS

  • Build an end-to-end machine learning project
  • Understand the business problem
  • Collect and clean data
  • Explore and analyze the dataset
  • Select an appropriate model
  • Train the model
  • Evaluate model performance
  • Improve the results
  • Document the project
  • Present findings professionally
  • Prepare the project for a portfolio

PRACTICE & LABS

  • Analyze e-commerce store performance
  • Forecast future sales
  • Analyze marketing campaigns
  • Classify customers by purchasing value
  • Predict customer churn risk
  • Recommend products based on customer behavior
06Deep Learning Foundations

TOPICS

  • Introduction to deep learning
  • Machine learning versus deep learning
  • Artificial neural networks
  • The perceptron
  • Neural network layers
  • Activation functions
  • Loss functions
  • Optimizers
  • Backpropagation explained simply
  • Build a neural network with TensorFlow or PyTorch
  • Train a neural network
  • Evaluate a neural network
  • Overfitting in deep learning
  • Dropout
  • Batch normalization
  • Saving and loading trained models

PRACTICE & LABS

  • Build a customer churn prediction model
  • Classify spam messages
  • Recognize handwritten digits
  • Identify high-value customers
  • Predict house prices with a neural network
07RNN, LSTM & Sequential Data

TOPICS

  • Introduction to sequential data
  • Why do we need RNNs?
  • How recurrent neural networks work
  • Limitations of traditional RNNs
  • Vanishing gradients
  • Introduction to LSTM
  • Differences between RNNs and LSTMs
  • Using LSTM models for text
  • Using LSTM models for time series
  • Representing text numerically
  • Text preprocessing
  • Tokenization
  • Building a simple text classifier
  • Introduction to the attention mechanism

PRACTICE & LABS

  • Perform sentiment analysis on text
  • Classify text messages
  • Forecast simple time-series data
  • Build a basic sequence understanding model
08Computer Vision

TOPICS

  • Introduction to computer vision
  • Understanding images as digital data
  • Working with images in OpenCV
  • Reading and modifying images
  • Image filtering
  • Edge detection
  • Image augmentation
  • Introduction to convolutional neural networks (CNNs)
  • Building an image classifier
  • Common computer vision model families
  • Object detection
  • Using YOLO for object detection
  • Training a custom computer vision model
  • Evaluating image model performance
  • Connecting a vision model to an app or API

PRACTICE & LABS

  • Classify images into two categories
  • Build a multiclass image classifier
  • Detect objects using YOLO
  • Explore facial-expression emotion classification
  • Build an image-based food or product suggestion prototype
09Transfer Learning & Embeddings

TOPICS

  • What is transfer learning?
  • Why use pre-trained models?
  • Feature extraction
  • Fine-tuning
  • Using pre-trained models for image classification
  • Binary image classification
  • Multiclass image classification
  • Improving performance with limited data
  • Transfer learning for NLP
  • The concept of embeddings
  • Converting text into vectors
  • Using embeddings for classification and search
  • Introduction to Hugging Face models

PRACTICE & LABS

  • Classify images with a pre-trained model
  • Fine-tune an image classifier
  • Use embeddings to classify spam messages
  • Analyze and classify Arabic tweets
  • Classify news articles by topic
10Transformer Models

TOPICS

  • Why transformers were introduced
  • RNN/LSTM versus transformers
  • Tokenization
  • Embeddings
  • Positional encoding
  • Attention mechanisms
  • Self-attention
  • Multi-head attention
  • Encoder architecture
  • Decoder architecture
  • Encoder-decoder architecture
  • Comparing BERT, GPT, and T5
  • Transformers for understanding text
  • Transformers for generating text
  • Fine-tuning for specific tasks
  • Using Hugging Face Transformers
  • Choosing the right model for the task
  • Evaluating transformer model performance

PRACTICE & LABS

  • Text classification with a transformer
  • Sentiment analysis of Arabic text
  • Text summarization
  • Question answering
  • Simple text generation
  • Compare outputs from different models
11LLM Engineering

TOPICS

  • Introduction to large language models (LLMs)
  • Comparing GPT, Llama, DeepSeek, and other model families
  • Understanding tokens
  • Context window
  • Temperature
  • Top-p
  • Maximum output tokens
  • Sampling strategies
  • System prompts
  • User prompts
  • Assistant responses
  • Prompt engineering
  • Designing effective prompts
  • Few-shot prompting
  • Safe, structured approaches to reasoning prompts
  • Structured outputs
  • Function calling and tool calling
  • Working with JSON outputs
  • Comparing model quality, cost, and latency
  • Running language models locally
  • Using Ollama or LM Studio
  • Cost optimization fundamentals
  • Managing hallucinations
  • Safety and responsible use of LLMs

PRACTICE & LABS

  • Build an AI assistant for user questions
  • Design reusable prompts for different tasks
  • Generate validated structured JSON output
  • Compare GPT, Llama, and DeepSeek on practical tasks
  • Run and test a local model
12RAG & Vector Databases

TOPICS

  • Retrieval-Augmented Generation (RAG) fundamentals
  • When to use RAG
  • Model-only answers versus source-grounded answers
  • Revisiting embeddings
  • Semantic search
  • Vector databases
  • Storing and retrieving vector representations
  • Document loading
  • Text splitting
  • Chunking strategies
  • Retrieval
  • Reranking
  • Building a RAG pipeline
  • Evaluating answer quality
  • Reducing hallucinations with RAG
  • Working with PDF, Word, and CSV documents
  • Building a chatbot for internal documents
  • Data protection and privacy

PRACTICE & LABS

  • Build a chatbot that answers questions from a PDF
  • Build an assistant grounded in company policies
  • Build semantic search over documents
  • Build a training assistant grounded in course material
  • Evaluate RAG answers for accuracy and grounding
13AI Agents & MCP

TOPICS

  • Understanding AI agents
  • Chatbots versus agents
  • Tools
  • Memory
  • Planning
  • Actions
  • Connecting LLMs to APIs
  • Introduction to the Model Context Protocol (MCP)
  • MCP clients and MCP servers
  • Resources, tools, and prompts in MCP
  • Building agents that perform practical tasks
  • Using multiple tools in one agent
  • Handling errors during execution
  • Logging and monitoring an agent
  • Operational boundaries of AI agents
  • Security when connecting agents to live systems

PRACTICE & LABS

  • Develop a weather assistant
  • Build a Telegram-based e-commerce sales agent
  • Build an appointment and booking workflow
  • Develop a customer support assistant
  • Explore safe integrations with financial systems
  • Build an API-connected agent with structured responses
14AI Deployment & MLOps Foundations

TOPICS

  • Why a model inside a notebook is not enough
  • Saving and loading machine learning models
  • Building APIs with FastAPI
  • Creating prediction endpoints
  • Testing APIs with Postman
  • Building a simple Streamlit interface
  • Introduction to Docker
  • Environment variables
  • Publishing projects to GitHub
  • Deployment fundamentals
  • Monitoring models after deployment
  • Logging
  • Error handling
  • Improving response times
  • Reducing operating cost
  • Introduction to MLOps practices

PRACTICE & LABS

  • Turn an ML model into an API
  • Build a Streamlit interface for predictions
  • Deploy a simple AI application
  • Document an API for a development team
15AI Systems for E-Commerce

TOPICS

  • Analyzing e-commerce store data
  • Analyzing customer behavior
  • Identifying best-selling products
  • Market basket analysis
  • Building related-product recommendation systems
  • Personalized product recommendations
  • Forecasting future sales
  • Predicting customer churn
  • Building an AI customer support assistant
  • Building an LLM-powered sales assistant
  • Using RAG for store policies and FAQs
  • Using computer vision for product suggestions
  • Creating APIs to integrate AI models with a store
  • Deploying AI features in an e-commerce store
  • Measuring AI impact on sales and customer experience

PRACTICE & LABS

  • Build an integrated AI system for an online store that includes:
  • Sales and customer analytics dashboard
  • Product recommendation model
  • Sales forecasting model
  • AI customer support assistant
  • APIs connecting models to the storefront
  • Final report with findings and performance indicators
16Optional: Reinforcement LearningOptional

TOPICS

  • Introduction to reinforcement learning
  • Supervised learning versus reinforcement learning
  • Agent
  • Environment
  • Reward
  • State and action
  • Markov decision processes
  • Q-learning
  • Policy and value functions
  • Using Gymnasium
  • Introduction to deep reinforcement learning
  • Limitations and challenges of reinforcement learning

PRACTICE & LABS

  • Train an agent for a Flappy Bird-style game
  • Solve navigation and maze tasks
  • Build a simple agent for Atari environments
  • Simulate an autonomous driving agent
  • Prototype robot arm control in simulation
  • Optimize delivery operations in a simulated environment
Photograph from a professional recognition event, provided by the academyFrom a professional recognition event

LEARN DIRECTLY WITH MOSTAFA

Learn with Eng. Mostafa Nabih

An AI and data solutions consultant with over 7 years of professional and training experience across software development, analytics, and data engineering. Holds Microsoft PL-300 and DP-700 certifications and was part of a team recognized as an African runner-up in an innovation competition.

Across all seven live sessions, Mostafa will guide hands-on development with Claude for requirements, architecture, and review, and Codex for implementing WordPress, web, native mobile, and API components. He will explain decisions, review generated code, run tests, and verify security before deployment.

Microsoft PL-300Microsoft DP-700AI & Data EngineeringTechnical Training
7+ YearsExperience
02Microsoft certifications
7Live build sessions

The photo illustrates the instructor’s professional experience and does not imply any endorsement by the organization pictured.

CAPSTONE PROJECT STUDIO

Choose the System You Want to Build

Three possible capstone themes sharing one adaptable architecture.

SELECTED CAPSTONE

AI Learning Platform

WordPress landing/CMS, learner web app, native mobile client, and a document-grounded RAG course assistant.

WordPressWeb AppMobileAI APIVPS

YOUR PORTFOLIO DELIVERABLES

Make Your Learning Visible

Target deliverables, subject to completing the practice and having the required services available.

⌘ GitHub Repository
W WordPress Site
▣ Web Dashboard
▥ Mobile Test App
{} FastAPI / RAG
⌁ VPS + HTTPS
☑ Tests & Documentation
↗ Portfolio Demo

FAQ

Before You Enroll

Clear answers about recorded learning, live classes, and project requirements.

What is the difference between recorded and live?

Recorded learning covers 16 progressive tracks across Python, ML, LLMs, RAG, agents, and more at your own pace. The 7 live classes with Mostafa focus on building one integrated WordPress + web + native + AI system and deploying it.

Will we actually build with Claude and Codex?

Yes. Claude and Codex are used hands-on throughout all seven sessions. Claude supports requirements, architecture, and review; Codex supports code generation, implementation, tests, and debugging. The instructor explains decisions and validates outputs rather than accepting generated code blindly.

Do I need coding experience?

The recorded path begins with Python fundamentals. Live classes move quickly, so completing the Python, Git, and API foundations before the build project is strongly recommended.

Will we build three unrelated apps?

No. You will build one system: WordPress for marketing/CMS, a web app for the user workspace, and a native mobile client, all connected to a shared backend and AI API.

Will I deploy to a real server?

The last class covers VPS, DNS, Docker, HTTPS, backups, and monitoring. A live deployment requires an available server, domain, and suitable access; third-party infrastructure charges may apply.

Do we publish to the app stores?

You will build a React Native/Expo app and learn testing and test builds. Store publication requires separate developer accounts and reviews; approval is not guaranteed.

Can seven sessions teach every AI topic?

No. The recorded curriculum provides depth, while the live sessions focus on delivering one integrated MVP. Your project scope depends on the practice completed between sessions.

Do Claude or Codex require paid access?

Access and usage limits depend on each provider’s current plans. Paid subscriptions and API credits should not be assumed included in tuition; required accounts and possible alternatives will be clarified before the build.

Will AI tools cost extra?

Local and free options are discussed where practical; hosted models, servers, domains, and app stores can incur extra charges unless explicitly included in enrollment terms.

When do the live sessions start, and how do I choose a cohort?

Select the 1 November 2026 or 1 December 2026 cohort before checkout. These are the live-cohort start dates with Mostafa; the recorded content is self-paced. If neither works, explore private training.

BUILD YOUR NEXT BIG PROJECT

Build, Integrate & Deploy Your AI Project.

Build technical depth with recorded learning. In live sessions with Mostafa, use Claude and Codex to deliver one integrated system, with code review and testing before deployment.

Deployment may involve separate VPS, domain, AI API, and app-store account charges.

Choose Your Cohort
Applied AI BootcampChoose Your Cohort

Want to receive push notifications for all major on-site activities?

✕