19 courses, grouped by topic and listed in learning order. Every course stands on its own — pick any one and start. All of them are written, with 398 lab source files behind them: every numeric claim here was produced by running code.
Number systems, memory hierarchy, networks, Word, Excel, pivot tables and dashboards. Ten programs re-compute every spreadsheet figure and every number-system conversion.
5 units + labControl flow, arrays, strings, pointers, structures and file handling. 15 runnable programs.
5 units + labSyntax, collections, files, exceptions, OOP, linked lists, stacks, queues. 18 experiments; 16 run, the two Tkinter programs syntax-checked only.
5 units + labHTML structure and forms, CSS layout and responsive design, JavaScript and the DOM, client-side validation, JSON and jQuery. 16 experiments, 184 assertions under jsdom.
Probability, distributions, correlation, regression, estimation and hypothesis testing. Formula sheet included.
5 units + labThe data science lifecycle, R and RStudio, dplyr and ggplot2, regression and clustering, ARIMA and Shiny. 18 R scripts, 14 with executed Python equivalents.
5 units + labNumPy arrays and broadcasting, Pandas Series and DataFrames, cleaning and feature engineering, reshaping and merging, matplotlib, Seaborn and Plotly. All 18 practicals run.
5 units + labStationarity, ACF and PACF, ARMA, ARIMA and SARIMA, model selection, prediction intervals, VAR and Granger causality, Kalman filtering and spectral analysis. Every one of the 13 experiments runs.
Three-schema architecture, ER modelling, normalization to 3NF, SQL joins and PL/SQL. Executable SQL labs.
5 units + labNoSQL and CAP, the BSON document model, CRUD and MQL, embed against reference, aggregation pipelines, indexing and replication. Sixteen of the twenty experiments execute through mongomock; the other four need a real server, and every mongosh script says it was not executed.
5 units + labHDFS and YARN, MapReduce, Hive and Pig, ingestion and serialization, HBase, ZooKeeper and Spark. 14 of 17 practicals run, including real Apache Spark, Avro and Parquet.
5 units + labService and deployment models, virtualization, cloud storage and warehouses, managed ML, deployment and monitoring. IAM evaluation, a real ETL and a real endpoint, all run locally.
Data warehousing and OLAP, preprocessing and similarity, Apriori and FP-Growth, ID3 and C4.5, K-Means and DBSCAN. Every hand trace re-executed in scikit-learn.
5 units + labML paradigms, preprocessing and evaluation, regression, classification and clustering. All 12 practicals run under scikit-learn.
5 units + labIntelligent agents, uninformed and informed search, CSPs, propositional and first-order logic, expert systems. 19 experiments; the Prolog is marked not executed, and five of the Python halves run as real logic programs.
5 units + labPerceptrons and activations, backpropagation, CNNs, RNNs and LSTM, attention and transformers, transfer learning and AI ethics. Ten of twelve experiments run against real MNIST, Fashion-MNIST, IMDb and real ImageNet weights.
5 units + labAmbiguity and regular expressions, tokenization, stemming and lemmatization, grammars and parsing, named entity recognition, embeddings, classification, RNNs and transformers. Eleven of fourteen experiments run against real NLTK corpora and real spaCy models.
BI concepts and the data warehouse, Power BI and Power Query, DAX, Tableau and LOD expressions, dashboard design. Every figure computed, tool click-paths marked not executed.
5 units + labThe data engineering lifecycle, architecture and distributed systems, MLOps fundamentals, deployment and CI/CD, monitoring and governance. Eleven of sixteen experiments run against real MLflow, git, DVC and Flask.
Fifty CSV datasets covering the methods these courses teach — each generated from a known truth, so you can score your answer rather than just produce one.
266 questions across those datasets, graded warm-up to stretch — every answer computed from the file rather than written from memory.
The official programme the courses were written to, kept as published. Its semesters, elective tracks and course numbers are the programme’s, not this site’s.
The programme’s structure as its documents set it out, with unit-level topics.
Thirty-three findings from checking the four official syllabus documents.
The week-by-week plan written for the original programme, with revision cycles and a progress checklist.
The official syllabus text, extracted verbatim from its four PDFs: part one, part two, part three and part four.