Data Science from Scratch: First Principles with Python

What it’s about

Data Science from Scratch teaches you data science by having you build the tools and algorithms yourself, from the ground up, in clear and readable Python. Rather than treating libraries like NumPy, scikit-learn, and pandas as magic black boxes, Joel Grus walks you through implementing linear algebra, statistics, probability, gradient descent, machine learning models, neural networks, deep learning, clustering, NLP, network analysis, and recommender systems by hand. Framed around the fictional 'DataSciencester' social network, the book grounds abstract concepts in concrete problems while developing the hacking skills and mathematical intuition that are at the core of doing real data science. By the end you understand not just how to use data science tools, but how and why they work.

The through-line

Who it’s for

An aspiring data scientist with some mathematical aptitude and programming skill who wants to genuinely understand how data science works, not just call library functions.

The problem

They need to learn the core algorithms, mathematics, and tools of data science well enough to actually do the work. They feel like an underachiever or impostor who can use libraries but doesn't truly understand what's happening under the hood.

The plan

1. Get comfortable with Python and the language features that matter for data science.
2. Build a foundation in linear algebra, statistics, and probability.
3. Learn to get, clean, explore, and manipulate real data.
4. Implement core machine learning models and evaluation techniques from scratch.
5. Advance to neural networks, deep learning, clustering, NLP, and recommender systems.

The payoff

You possess a solid understanding of the fundamentals of data science. · You can build, train, and evaluate models while understanding how they work. · You can confidently use production libraries because you know what they do under the hood.

See our guide

• Practice Business Intelligence & Data Science →
• The Four-S Discipline — One System, Four Capabilities →
• AI Safety, Governance & Alignment Practice (FIELD GUIDE (ALL LEVELS)) →
• Choose And Apply The Right Analytical Method →
• Build The Data And AI Systems Behind Analytics →
• Ship Analytics, Don't Just Prototype It →
• Get The Business To Actually Use It →
• Build A Model That Holds Up Outside Your Sample →
• Get The Data Trustworthy First →

Related profiles we’ve built

• A Practical Guide To Conjoint Analysis →
• Analytics at Work: Smarter Decisions, Better Results →
• Big Data: A Revolution That Will Transform How We Live, Work, and Think →
• Big Data_ A Very Short Introduction (Very Short Introductions) →
• Business Adventures →
• Business Intelligence Guidebook: From Data Integration to Analytics →
• Data Mining for Business Analytics: Concepts, Techniques, and Applications →
• Data Warehouse and Data Mining →

Additional reading

• Python for Data Analysis · Wes McKinney

  Recommended as the best way to learn the pandas library, which is the primary Python tool for cleaning, munging, and manipulating data in real-world applications.

• The Elements of Statistical Learning · Jerome H. Friedman, Robert Tibshirani, and Trevor Hastie

  Cited as a canonical (though math-heavy) textbook for gaining a deeper understanding of machine learning concepts.

• Introduction to Probability · Charles M. Grinstead and J. Laurie Snell

  Recommended as a good, freely available textbook for readers who want to build a stronger foundation in probability theory.

• scikit-learn documentation · N/A

  The book positions scikit-learn as the go-to library for production machine learning in Python, and its documentation is highlighted as a key resource for practical application.

• Deep Learning · Ian Goodfellow, Yoshua Bengio, and Aaron Courville

  Presented as the canonical textbook on deep learning, for readers who want a mathematically rigorous and comprehensive understanding of the field.

• pandas documentation · N/A

  The book repeatedly contrasts its from-scratch data manipulation with the more efficient and powerful methods available in the pandas library, making its documentation essential for practical work.

Featured in these guides

• Ship Analytics, Don't Just Prototype It →Cited source
• Practice Business Intelligence & Data Science →Cited source
• AI Safety, Governance & Alignment Practice (FIELD GUIDE (ALL LEVELS)) →Cited source
• Get The Business To Actually Use It →Cited source
• Build A Model That Holds Up Outside Your Sample →Cited source
• Choose And Apply The Right Analytical Method →Cited source
• Build The Data And AI Systems Behind Analytics →Cited source
• Get The Data Trustworthy First →Cited source