Ship working systems
Practitioner track
Weighted towards code, metrics, failure modes and deployment. You get the maths you need to debug a model, plus the engineering the textbooks leave out.
Best if: Building models or LLM applications in industry.
- 01
Start Here
How to use VibeFormer, how the three learning tracks differ, and the notation conventions used throughout.
5 of 5 lessons on this track
- 02
Programming, Data Structures and Algorithms
Python for data work, then the full DSA syllabus: complexity, linear structures, trees, hashing, sorting, searching and graph algorithms.
27 of 27 lessons on this track
+21 more
- 03
Databases and Data Warehousing
Relational modelling, algebra, SQL, normalisation and indexing, then OLAP, dimensional schemas and the preprocessing pipeline.
23 of 23 lessons on this track
- The Entity–Relationship Model
- The Relational Model and Keys
- Relational Algebra
- Tuple Relational Calculus
- SQL Fundamentals
- SQL Joins and Aggregation
+17 more
- 04
Optimisation Algorithms
Linear and integer programming, duality, first-order and interior-point methods, metaheuristics and Bayesian optimisation.
21 of 27 lessons on this track
- The Optimisation Landscape
- Formulating an Optimisation Problem
- Linear Programming
- The Simplex Method
- Linear Programming Duality
- Sensitivity and Post-Optimality Analysis
+15 more
- 05
Machine Learning Foundations
The concepts every algorithm shares: risk minimisation, generalisation, the bias–variance trade-off, validation and metrics.
18 of 20 lessons on this track
- What Machine Learning Actually Isready
- Formulating a Learning Problemready
- Empirical Risk Minimisationready
- Generalisation, Overfitting and Underfittingready
- The Bias–Variance Trade-offready
- Train, Validation and Test Splitsready
+12 more
- 06
Supervised Learning
Every supervised algorithm on the syllabus, each derived from its objective, traced on small numeric data, then coded from scratch.
24 of 25 lessons on this track
- Simple Linear Regression
- Multiple Linear Regression
- Regression Assumptions and Diagnostics
- Polynomial and Basis Expansion Regression
- Ridge Regression
- Lasso and Elastic Net
+18 more
- 07
Unsupervised Learning
Clustering, dimensionality reduction, association rules and anomaly detection, each traced numerically and derived where it matters.
22 of 26 lessons on this track
- The Unsupervised Landscape
- Distance and Similarity Measures
- k-Means Clustering
- Initialisation and Choosing k
- k-Medoids and PAM
- Agglomerative Hierarchical Clustering
+16 more
- 08
Neural Networks and Deep Learning
Backpropagation derived and computed by hand, then optimisers, CNNs, RNNs, autoencoders, VAEs, GANs and diffusion models.
36 of 38 lessons on this track
- From Perceptron to Deep Networks
- Forward Propagation
- Backpropagation
- Backpropagation: A Complete Numeric Example
- Activation Functions
- Loss Functions
+30 more
- 09
Natural Language Processing
Tokenisation and n-grams through embeddings, attention and the complete transformer, with shapes traced end to end.
30 of 32 lessons on this track
- The NLP Pipeline
- Text Normalisation
- Regular Expressions for Text
- Stemming and Lemmatisation
- n-Gram Language Models
- Smoothing
+24 more
- 10
Large Language Models
How modern LLMs are built, aligned, decoded, evaluated, served and turned into agents — with the mechanics, not the hype.
27 of 32 lessons on this track
- What an LLM Actually Is
- Decoder-Only Architecture in Detail
- Pretraining Objectives
- Pretraining Data Pipelines
- Scaling Laws
- Compute and Token Economics
+21 more
- 11
Fine-Tuning and Model Adaptation
Full fine-tuning, LoRA and QLoRA, quantisation, distributed training and data curation — with the maths behind each.
23 of 26 lessons on this track
- Prompt, RAG or Fine-Tune?
- Full Fine-Tuning
- Catastrophic Forgetting
- Instruction Tuning
- Dataset Formats and Chat Templates
- Parameter-Efficient Fine-Tuning: The Landscape
+17 more
- 12
Retrieval-Augmented Generation
Chunking, embeddings, vector indexes, hybrid retrieval, reranking and evaluation, through to agentic and graph RAG.
26 of 28 lessons on this track
- Why RAG Exists
- RAG Architecture
- Document Ingestion and Parsing
- Chunking Strategies
- Embedding Models
- Similarity Metrics for Retrieval
+20 more
- 13
Time Series Analysis
Stationarity, ACF/PACF, the ARIMA family, exponential smoothing, state-space models and deep forecasting.
14 of 15 lessons on this track
- Components of a Time Series
- Stationarity
- Unit Root Tests
- ACF and PACF
- Autoregressive Models
- Moving Average Models
+8 more
- 14
MLOps and Responsible AI
Getting models into production and keeping them honest: versioning, monitoring, drift, fairness, privacy and governance.
15 of 16 lessons on this track
- Framing an ML Problem
- Data and Model Versioning
- Experiment Tracking
- Model Registry and Packaging
- Deployment Patterns
- Monitoring and Drift Detection
+9 more