Beginner to Practitioner - 12 Weeks

AI Fundamentals

Deep dive into artificial intelligence and machine learning. Build intelligent systems using TensorFlow, PyTorch, and modern MLOps practices.

12 Weeks Duration
300+ Topics
6 Projects
Enroll Now View Curriculum

Program Overview

AI Fundamentals is a 12-week hands-on program designed to take you from beginner to practitioner in artificial intelligence and machine learning.

This 12-week program provides a structured, hands-on path from beginner to practitioner in artificial intelligence and machine learning. Students progress through four carefully sequenced phases - building mathematical foundations, mastering classical ML algorithms, diving into deep neural networks, and finally applying skills to production-ready systems.

Phase Weeks Focus Area Outcome
Phase 1 Weeks 1-3 Foundations: Python, Maths, Data Preprocessing Confident with data pipelines and linear algebra
Phase 2 Weeks 4-6 Core ML: Regression, Classification, Unsupervised Can build and evaluate classical ML models
Phase 3 Weeks 7-10 Deep Learning: DNNs, CNNs, RNNs, Transformers Trains deep models in TensorFlow and PyTorch
Phase 4 Weeks 11-12 Applied AI: MLOps, Deployment, Capstone Project Ships a live model with portfolio documentation

Weekly Curriculum

Click any week to expand topics and lab assignment.

W1

Phase 1 - Foundations

Foundations of AI & Python for ML

Topics Covered

History of AI Types of ML Python refresher
NumPy Pandas Matplotlib
Jupyter Notebooks Data types Data structures

Lab / Deliverable

Set up your ML dev environment and perform exploratory data analysis (EDA) on a real dataset.

W2

Phase 1 - Foundations

Mathematics for Machine Learning

Topics Covered

Vectors & matrices Matrix multiplication Eigenvalues
Derivatives & gradients Chain rule Probability basics
Distributions Bayes theorem Statistical inference

Lab / Deliverable

Implement gradient descent from scratch in NumPy to understand parameter optimization.

W3

Phase 1 - Foundations

Data Preprocessing & Feature Engineering

Topics Covered

Handling missing values Encoding categoricals Normalization & scaling
Feature selection Dimensionality reduction Train/val/test splits
Data leakage scikit-learn Pipelines Outlier detection

Lab / Deliverable

Build a full preprocessing pipeline for a messy Kaggle dataset from scratch.

W4

Phase 2 - Core ML

Supervised Learning I - Regression

Topics Covered

Linear regression Polynomial regression Cost functions
Gradient descent R-squared, RMSE, MAE Ridge (L2)
Lasso (L1) Elastic Net Bias-variance tradeoff

Lab / Deliverable

Project 1: House price prediction model with full evaluation and regularization tuning.

W5

Phase 2 - Core ML

Supervised Learning II - Classification

Topics Covered

Logistic regression Decision trees Random forests
Gradient boosting XGBoost SVM
k-NN Confusion matrix Precision / Recall / F1
ROC-AUC Cross-validation

Lab / Deliverable

Project 2: Binary and multi-class classification on a medical or financial dataset.

W6

Phase 2 - Core ML

Unsupervised Learning

Topics Covered

K-Means clustering Hierarchical clustering DBSCAN
Gaussian mixture models PCA t-SNE
UMAP Autoencoders (intro) Anomaly detection

Lab / Deliverable

Customer segmentation using K-Means with PCA visualisation of cluster structure.

W7

Phase 3 - Deep Learning

Deep Neural Networks - Foundations

Topics Covered

Perceptrons Multi-layer perceptrons Activation functions
Backpropagation Optimizers (SGD, Adam) Weight initialization
Batch normalization Dropout Overfitting & regularization

Lab / Deliverable

Build and train a neural network in pure NumPy, then replicate in TensorFlow/PyTorch.

W8

Phase 3 - Deep Learning

TensorFlow & PyTorch in Practice

Topics Covered

TensorFlow 2.x / Keras API PyTorch tensors Autograd
Custom training loops DataLoaders GPU acceleration
Model saving & loading TensorBoard Callbacks & hooks

Lab / Deliverable

Train the same image classifier in both frameworks and compare APIs, speed, and ergonomics.

W9

Phase 3 - Deep Learning

Convolutional Neural Networks (CNNs)

Topics Covered

Convolution operation Pooling layers CNN architectures
VGG & ResNet Transfer learning Fine-tuning
Data augmentation Object detection (intro) Grad-CAM visualisation

Lab / Deliverable

Project 3: Image classifier using transfer learning from ResNet, fine-tuned on a custom dataset.

W10

Phase 3 - Deep Learning

Sequence Models & NLP Foundations

Topics Covered

Recurrent neural networks LSTMs & GRUs Vanishing gradients
Word embeddings Word2Vec / GloVe Attention mechanism
Transformer architecture BERT / GPT overview Hugging Face intro

Lab / Deliverable

Sentiment analysis with LSTM, then fine-tune a BERT model using Hugging Face Transformers.

W11

Phase 4 - Applied AI

MLOps & Model Deployment

Topics Covered

ML project lifecycle MLflow experiment tracking Hyperparameter tuning
Docker for ML FastAPI model serving REST API design
Model versioning Data drift monitoring CI/CD for ML

Lab / Deliverable

Project 4: Package and deploy a trained model as a live REST API with MLflow tracking and Docker.

W12

Phase 4 - Applied AI

Capstone - Real-World AI Project

Topics Covered

Problem scoping Dataset sourcing Full ML pipeline
Model selection & tuning Fairness & bias audit Responsible AI principles
Technical presentation Portfolio writeup

Capstone Project

Choose from: Fraud Detection System, Recommendation Engine, Medical Image Classifier, or NLP Chatbot. Present a live demo.

Learning Outcomes

Skills you'll have by the end of the program.

ML Fundamentals

Understand and explain supervised, unsupervised, and deep learning from first principles

Python & Libraries

Use NumPy, Pandas, Matplotlib, and scikit-learn confidently for any ML task

Framework Fluency

Write production-grade models in both TensorFlow/Keras and PyTorch

Computer Vision

Build image classifiers and apply transfer learning with pre-trained CNNs

NLP & Transformers

Implement sequence models and fine-tune BERT-class models with Hugging Face

Model Deployment

Serve models via REST APIs using FastAPI and Docker, with MLflow tracking

Responsible AI

Audit models for bias and apply ethical AI frameworks to real-world systems

Portfolio Projects

Present 6 completed, documented projects to potential employers

Assessment & Grading

30%

Weekly Labs (Weeks 1-10)

Practical coding labs submitted after each session

25%

Mid-Program Projects (x2)

House price prediction + Classification project (Weeks 4-5)

20%

Deep Learning Project

CNN image classifier with transfer learning (Week 9)

10%

MLOps Deployment Lab

Live REST API deployment with MLflow (Week 11)

15%

Capstone Project

End-to-end AI system with live demo and portfolio writeup

Tools & Technologies

Languages

Python 3.10+

Core Libraries

NumPy, Pandas, Matplotlib, Seaborn, scikit-learn

Deep Learning

TensorFlow 2.x, Keras, PyTorch, Hugging Face Transformers

MLOps

MLflow, Docker, FastAPI, Git, GitHub Actions

Environments

Jupyter Notebook, Google Colab, VS Code

Datasets

Kaggle, UCI ML Repository, Hugging Face Datasets

Prerequisites & Requirements

This program is designed for beginners. No prior ML experience is required.

Basic programming experience in any language (Python experience is a plus)

A laptop with internet access (cloud GPU access provided for deep learning modules)

8-10 hours per week available for study, labs, and projects

Curiosity and willingness to engage with mathematical concepts at an introductory level

Ready to Build Intelligent Systems?

Join the next generation of AI engineers. Go from zero ML knowledge to deploying production models in 12 weeks.

Enroll Now View All Courses