Full transparency · Open methodology · Open source
Open Source Textbooks
An Introduction to Statistical Learning (ISL) free pdf
Gareth James, Daniela Witten, Trevor Hastie, Robert Tibshirani
The primary reference for Levels 3 and 6. Covers regression, classification, resampling, tree-based methods, and unsupervised learning with R labs throughout. Used as the pedagogical backbone of this curriculum.
Primary reference for Level 4 (The Tidyverse). Covers data import, transformation, visualisation, and communication using the tidyverse. The definitive guide to modern R practice.
Forecasting: Principles and Practice (FPP3) free online
Rob J Hyndman, George Athanasopoulos
Primary reference for Level 5 (Time Series). Covers time series decomposition, exponential smoothing, ARIMA models, and forecast evaluation. Essential reading for anyone working with market data.
David Diez, Mine Çetinkaya-Rundel, Christopher Barr
Supplementary statistics reference. Used as a gentler introduction to probability and inference concepts before ISL. Freely available and community-maintained.
The definitive guide to pandas, NumPy, and the Python data ecosystem. Primary reference for Levels 2–4 of the Python Lab curriculum. Covers DataFrames, time series, and data wrangling in depth.
Primary reference for Level 5 (Machine Learning). Covers classification, regression, cross-validation, pipelines, and model evaluation. The most widely-used ML library in production Python environments.
Core reference for Level 1 and Level 3. NumPy underpins nearly all scientific Python computing — arrays, linear algebra, random number generation, and vectorized operations.
Reference for Level 5 ARIMA and time series modeling. Provides ARIMA, OLS regression, statistical tests, and forecasting tools. Essential for quantitative finance work.
The engine that runs real Python code (including NumPy, pandas, scikit-learn) directly in your browser with no server. A remarkable piece of open-source engineering that makes the Python Lab possible.
The engine that runs real R code directly in your browser with no server required. This is what makes the R Lab possible without backend infrastructure. A remarkable piece of open-source engineering.
This platform was built using Claude (Anthropic) as the primary engineering tool.
Claude authored the majority of the HTML, CSS, JavaScript, and R curriculum structure for this project — including the R Lab interface, all 23 trial definitions, validation logic, the JS simulation engine, and this references page. The AI model used was Claude Sonnet 4.6.
All design decisions, content review, financial framing, and deployment were made by the project owner. Claude was the engineer; the human was the architect.
We believe transparency about AI-assisted development matters — especially on a platform that teaches data science. If you're curious about how this was built, the methodology is the same as the curriculum: iterative, tool-assisted, and documented.