Python & Jupyter
Core language and notebooks.
Skills you'll build
Python, NumPy & pandasProbability & statisticsFeature engineeringRegression & classificationModel evaluation & validationUnsupervised learningDeep learning with PyTorchNLP & LLM fundamentalsModel deployment (FastAPI)MLOps basics (MLflow)
Who this is for
Data analysts, developers, and STEM graduates who want to build machine learning models and move into data science roles.
What you'll need
Everything to have in place before day one.
Prerequisites
- Basic Python (variables, loops, functions)
- Secondary-school maths; we refresh algebra and statistics
- Data Analyst course or equivalent experience recommended
Equipment & time
- A laptop with at least 8 GB RAM (16 GB recommended)
- Stable internet for live sessions and cloud notebooks
- 10–12 hours per week for classes, labs, and projects
- Free Google Colab and Kaggle accounts for GPU labs
Tools you'll use
The same tools used by teams in the industry today.
pandas, NumPy, matplotlib
Data wrangling and visualisation.
scikit-learn
Classical machine learning.
PyTorch
Deep learning models.
MLflow
Experiment tracking and model registry.
FastAPI & Docker
Serve and package models.
Course outline & syllabus
Six phases, taken in order. Each ends with a practical lab so you apply what you learn straight away.
Phase 1: Python & Data Wrangling
Foundations- Python refresher: data structures, functions, and modules
- NumPy arrays and vectorised operations
- pandas for cleaning, joining, and reshaping data
- Exploratory data analysis and visualisation
- Git and reproducible notebooks
Phase 2: Statistics & Probability
Essential- Probability, distributions, and sampling
- Hypothesis testing and confidence intervals
- Linear algebra and calculus intuition for ML
- Experiment design and A/B testing
- Bias, variance, and common statistical traps
Phase 3: Machine Learning
Core- Supervised learning: linear and logistic regression, trees, ensembles
- Feature engineering and handling imbalanced data
- Cross-validation, metrics, and hyperparameter tuning
- Unsupervised learning: clustering and dimensionality reduction
- Model interpretability and fairness
Phase 4: Deep Learning & NLP
Advanced- Neural networks and training with PyTorch
- Computer vision with convolutional networks
- NLP: text processing, embeddings, and transformers
- Working with large language models and retrieval (RAG)
- Transfer learning and fine-tuning
Phase 5: Deployment & MLOps
Production- Serving models with FastAPI
- Packaging with Docker and deploying to the cloud
- Experiment tracking and model registry with MLflow
- Monitoring drift and retraining
- Responsible AI and data protection
Phase 6: Capstone & Career
Career- End-to-end ML project from problem framing to deployed API
- Write a technical report and model card
- Present results and trade-offs to a panel
- Certification paths: Azure DP-100, AWS ML Engineer Associate
- Job search: DS portfolio, take-home tasks, and interview practice
Portfolio projects
Build work you can show hiring managers, not just theory.
Churn prediction model
- Feature engineering and model comparison
- Business impact of precision vs recall
Customer segmentation
- Clustering on behavioural data
- Actionable segment profiles
NLP text classifier
- Transformer fine-tuning
- Evaluation on real-world text
Deployed ML API
- FastAPI + Docker deployment
- MLflow tracking and monitoring plan
Keep learning after the course
Your ongoing syllabus: what to study next as you grow in the role.
Deepen ML expertise
- Time-series forecasting
- Recommender systems
- Causal inference and uplift modelling
Go further with AI
- LLM applications, agents, and evaluation
- GPU training and accelerated computing
- Fine-tuning and serving open-source models
Production excellence
- Feature stores and MLOps platforms
- Kaggle competitions to sharpen skills
- Reading and reproducing research papers
Certifications & resources
Ready to start Data Science?
Register your interest and we'll recommend the right starting phase for your background and goals.