Beginner Level — Python Foundations & Core Machine Learning Algorithms
Goal:
To build Python fluency, get comfortable with Jupyter, and learn to manipulate and visualize
datasets and to cover ML fundamentals, math intuition, Sklearn implementations, and
evaluation metrics.
Introduction to Jupyter Notebook, Python essentials, and NumPy arrays for numerical
operations.
Loading CSV files, exploring datasets, filtering, grouping, merging, and handling missing
values.
Creating plots with Matplotlib, Seaborn, and interactive charts with Plotly to explore and
understand data.
EDA on a dataset. Clean, visualize, and summarize findings in a Jupyter notebook.
Overview of ML types (supervised vs unsupervised), concept of features/labels, train/test
split, and the ideas of overfitting & underfitting.
Mathematical intuition (equations, gradient descent), hands-on implementation from
scratch in Python, and with sklearn. Model evaluation with MSE, accuracy, and
confusion matrix.
Bayes’ theorem and Naive Bayes for classification (Play Tennis dataset, sentiment
analysis using Bag of Words). Decision Tree Classifier with sklearn, interpreting splits
and visualization.
Project on a Kaggle dataset. Preprocessing data, training multiple models, and
evaluating results.
Expert Track — Real-World ML Workflow & Feature Engineering
Goal:
To introduce ensemble models, feature engineering, evaluation metrics, and structured
ML projects.
Practical methods for transforming data: one-hot encoding, label encoding, scaling, Bag
of Words, and TF-IDF for text datasets.
Working with imbalanced data (Credit Card Fraud / Churn dataset), using
cross-validation, precision/recall, F1-score, and ROC curves to evaluate models.
Introduction to ensemble methods. Building reproducible ML projects with modular code,
sklearn pipelines, and version control.
End-to-end project. Handling messy data, applying feature engineering, training multiple
models, and evaluating with advanced metrics.
Pre-requisite: Basics of programming and maths
Course duration: 3 Months (Sep 2025 – Dec 2025)
Frequency: 2 sessions/week × 2 hrs
Marking Scheme (tentative)
Assessments: 2×30 = 60
Final Project: 40
Total Marks: 100 (60+40)
About the Instructor
Asma Khalil
Asma Khalil is an educator with a strong background in programming and applied AI. With experience in mentoring students, coordinating academic projects, and working on practical AI applications, she is passionate about making complex topics easy to understand. She brings a balance of technical knowledge and teaching skills, helping learners build confidence while exploring programming and machine learning.
Electronic Engineering
Teaching & Mentorship: Taught C++ programming fundamentals, OOP, Signals & Systems, Logic Design, and Computer Communication & Networking to undergraduate students. Guided final-year projects with a focus on applying AI and automation concepts.
Applied AI Exposure: Gained hands-on experience with machine learning, NLP, and computer vision techniques while working with an Austrian start-up. Worked on projects such as anomaly detection in traffic signs.
Industry & Training Work: Worked with Mindrift as an AI Tutor, evaluating and refining AI-generated responses to improve model performance.
Academic Coordination & Quality Assurance: Served as final-year project coordinator and member of the QA committee, providing guidance, feedback, and organizational support to the department.
Python programming, Machine Learning, Prompt Engineering, LLMs, Academic Mentorship & Research Guidance
- Developed dashboards in Finance and Marketing sector and optimize them.
- Completed Heart Disease Prediction Model using machine learning.
- Conducted multiple freelancing projects in data analysis on Upwork.
- Currently concluding Data Analytics course with AI integration