Python Course

Week 1: The Data Toolkit

The first week’s goal is to get everyone comfortable with the fundamental tools for handling and manipulating data in Python.

Day 1: Welcome to AI & Your Workspace

Briefly explain the difference between AI, Machine Learning, and Data Analysis. Outline the course journey.

Guided setup of the programming environment: Install Python, Jupyter Notebook, and essential libraries like Pandas, NumPy, and Matplotlib.

Day 2: Python Crash Course for Data

Python fundamentals: variables, data types (lists, dictionaries), and why they’re important for data.

 Write simple Python scripts in a Jupyter Notebook or VS Code to solve small data manipulation puzzles.

Day 3: Control Your Code with Functions & Loops

How to make code reusable (functions) and automate repetitive tasks (loops).

Write a function to clean a piece of text. Use a loop to process a list of numbers and many more such puzzles, to get you a feel for loops

Day 4: NumPy for High-Speed Numerical Operations

Introduce NumPy arrays and why they are much faster for numerical operations than standard Python lists.

Create NumPy arrays, perform mathematical operations, and practice selecting specific data using slicing.

Day 5: Pandas for Data Manipulation

Introduction to the DataFrame, the single most important tool in data analysis.

 Load a CSV file (e.g., a simple sales dataset) into a Pandas DataFrame. Use .head(), .info(), and .describe() to get a first look at the data.

Week 2: Data Exploration & Storytelling

This week is about taking raw data, cleaning it, exploring it to find insights, and communicating those insights through visualisations.

Day 6: The Art of Cleaning Data

Discuss why data cleaning is the most critical and time-consuming part of data analysis. Cover how to handle missing values and duplicates.

Take a messy dataset and clean it. Use Pandas functions like .isnull(), .dropna(), and .fillna().

Day 7: Finding Insights with EDA

What is Exploratory Data Analysis (EDA)? Using grouping and aggregation to ask questions of your data.

Use groupby () and aggregation functions (.mean(), .sum()) to find key business insights from a dataset.

Day 8: Basic Data Visualisation

The power of telling stories with charts. Introduction to Matplotlib and Seaborn.

Create fundamental plots: bar charts, histograms, and line plots to visualise trends and distributions.

Day 9: Advanced Visualisation for Deeper Relationships

How to visualise the relationship between two or more variables using scatter plots and heatmaps.

Use Seaborn to create scatter plots (to see correlations) and heatmaps (to visualize matrices).

Day 10: Mini-Project 1: Full EDA on a Real Dataset

Combining the week’s skills.

Students are given a new dataset (e.g., the Titanic dataset) and perform a complete EDA, from cleaning and analysis to creating insightful visualizations.

Day 11: Building Your Professional Portfolio with GitHub

Establish Your Professional Profile: You’ll start by understanding why GitHub is essential for your career as a data professional. Then, you’ll create your account, optimize your profile with a professional bio and headshot, and learn the difference between Git (local version control) and GitHub (your cloud-based portfolio).

Create Your Portfolio & Upload Your First Project: You will create your first public repository, which will serve as your data science portfolio. You’ll then directly upload your completed Exploratory Data Analysis (EDA) project notebook (.ipynb file) to showcase your practical skills.

Document Your Work with a High-Quality README: You’ll learn to craft a professional README.md file, the front page of your project. Using Markdown, you’ll clearly document the project’s overview, the dataset used, your key findings, and the tools you utilized, turning a simple code upload into a compelling portfolio piece for recruiters.

Week 3: Building Predictive Models

This week demystifies machine learning. Students will learn to train models that can make predictions based on data.

Day 12: Machine Learning Fundamentals

Supervised vs. Unsupervised learning. The importance of splitting data into “training” and “testing” sets.

Use Scikit-learn’s train_test_split to divide a dataset.

Day 13: Predicting Numbers with Linear Regression

 Explain the simple intuition behind predicting continuous values (like a house price or a student’s score).

 Build, train, and evaluate a simple linear regression model.

Day 14: Predicting Categories with Logistic Regression

How to predict a “yes/no” or “spam/not spam” outcome. Introduce the confusion matrix to evaluate performance.

 Build a classification model to predict customer churn or loan default.

Day 15: Powerful Classification with Tree-Based Models

Concept: The intuition behind Decision Trees and the “wisdom of the crowd” concept for Random Forests.

Practical: Build and visualize a Decision Tree. Then, use a Random Forest and compare its superior accuracy.

Day 16: Finding Hidden Groups with Clustering

 Introduction to unsupervised learning with K-Means clustering for finding patterns without pre-existing labels.

Use K-Means to segment customers into distinct groups based on their purchasing behavior.

Week 4 & Beyond: The AI Frontier

The final phase pushes into modern AI, focusing on Large Language Models (LLMs) and building an autonomous AI agent.

Day 16-17: Enhancing Model Performance

Feature Engineering (creating better input data) and Hyperparameter Tuning (finding the best model settings).

Create new features from existing data. Use GridSearchCV to automatically find the best parameters for a model.

Day 18-19: Mini-Project 2: End-to-End Prediction

Consolidate all machine learning skills.

Students choose a dataset and build a complete predictive model, from EDA and feature engineering to model training and evaluation.

Day 20: Intro to Deep Learning & NLP

A high-level, intuitive overview of neural networks and how machines process text    (Natural Language Processing).

Build a very simple neural network with Keras/TensorFlow. Use a library to perform basic text cleanup.

Day 21: The Power of LLMs & Prompt Engineering

What are LLMs (like Gemini, GPT)? The critical skill of prompt engineering.

Interact with an LLM API (e.g., Google’s Gemini API). Practice writing effective prompts to solve various tasks.

Day 22-23: Building Your Own Q&A Bot with RAG

What is Retrieval-Augmented Generation (RAG)? How to make an LLM answer questions using your own private documents.

Use a framework like LangChain to build a simple Q&A bot that can read a PDF or text file and answer questions about its content.

Day 24: Understanding AI Agents

The difference between a simple chatbot and an AI agent that can reason, plan, and use tools (like a calculator or web search).
Use a pre-built agent from LangChain to see how it can use tools to answer complex, multi-step questions.

Day 25-28: Final Project: Build Your First AI Agent

The capstone project.

Students get dedicated lab time to design and build a simple AI agent. Example: An agent that can analyse a simple CSV file you provide by writing and executing its own Python code to answer your questions. The instructor provides 1-on-1 support.

Day 29: Portfolio & Career Prep

How to showcase your projects on GitHub and structure a data-focused resume.

Students give a short presentation of their final agent. Peer review and feedback session.

Day 30: Your Future in AI & Final Q&A

Discuss specialisations (e.g., Computer Vision, MLOps), resources for continuous learning, and industry trends.