Best Technical Books: 36 Picks Recommended by CEOs

Ian Goodfellow, Yoshua Bengio and Aaron Courville start with linear algebra, probability and numerical computation, then teach the deep learning methods used in industry, from convolutional networks to sequence models. Elon Musk calls it the one comprehensive book on the subject, and Nassim Nicholas Taleb praises its clear math.

Harold Abelson and Gerald Jay Sussman’s MIT textbook teaches programming fundamentals in Scheme: recursion, abstraction, modularity and finally the interpreters and compilers that run languages. Patrick Collison lists it, and Santiago Basulto calls it the place to start on fundamentals.

Andreas M. Antonopoulos gives developers and engineers the cryptographic and technical foundations of bitcoin, including keys, addresses, wallets, transactions and the peer-to-peer network, with code snippets throughout. Auston Bunsen relied on its detailed chapter on wallet addresses for his company’s work.

Sebastian Raschka has readers code a large language model without LLM libraries, first a base model, then a text classifier, then a chatbot that follows instructions, all running on a laptop and assuming intermediate Python. Lex Fridman recommends it to beginners and experts alike.

Thomas M. Cover and Joy A. Thomas teach information theory through mathematics, physics and statistics, covering entropy, data compression, channel capacity and rate distortion, with problem sets closing each chapter. Eric Weinstein has it on his list.

Martin Fowler catalogs refactorings, each with motivation, mechanics and examples, and shows how to detect code smells and back changes with tests, now with JavaScript samples. David Heinemeier Hansson returns to it again and again, one of just two programming titles he rereads.

Kent Beck writes down as named patterns the Smalltalk habits that seasoned developers tend to learn by trial and error, covering methods, messages, state, collections and formatting. David Heinemeier Hansson names it the programming book he loves most, still relevant after two decades.

Peter Norvig teaches advanced Common Lisp by rebuilding real artificial intelligence programs, with chapters on debugging, efficiency and object-oriented programming aimed at building large, robust software. Patrick Collison has it on his bookshelf list.

Brian W. Kernighan and Rob Pike teach Unix from the shell and file system through grep, sed, AWK and make to writing C programs and building a small language with yacc. Tim O’Reilly admired both its tools philosophy and the clarity of its prose.

Ben Rapien distills 45 years with press brake tooling into a practical guide to forming sheet metal: choosing tools, checking machine requirements, calculating blank sizes and planning bend sequences, with the engineering simplified for operators. Carl Bass has recommended it.

The International Code Council spells out the technical rules for structure, plumbing, mechanical systems, fuel gas and electrical installations in houses and townhouses, down to seismic maps and header tables. Alan Pierce uses it for an overall grasp of modern Western building.

Steve Chappell’s workshop manual teaches the joinery, design and construction of traditional timber frames built on pegged mortise and tenon joints, with engineering specifications updated from new test data. Alan Pierce values it for rare joinery techniques behind houses that lasted centuries.

Bruce Schneier’s reference takes programmers from cryptographic protocols and techniques to the inner workings of algorithms such as RSA and DES, with source code in C and advice on random numbers and key security. Dominic Steil recommends it.

Arvind Narayanan and three co-authors lay the groundwork for engineering secure software on the Bitcoin network, covering mining, security and anonymity, with lecture slides and programming assignments online. Antonio Eram named it among his current reads.

Steve McConnell’s handbook on software construction gathers practical technique for coding, debugging, unit testing and integration, from naming variables to deciding when a subroutine pays off. Santiago Basulto places it beside SICP as a starting point for programmers moving into business.

Pete Goodliffe’s guide centers on the daily practice of writing excellent code, as its subtitle states, rather than on any single language. Lewis Smith, a developer, describes it as the quickest way for a junior programmer to become a senior one.

Rob Thallon’s visual handbook shows the hands-on detail of building a wood frame house, with drawings covering foundations, subfloors, wall framing, roof flashing and stairs, plus advanced framing and moisture control. Alan Pierce recommends it to anyone starting in construction or engineering.

Douglas Crockford, who specified the JSON format, trims JavaScript down to a dependable subset, covering objects, functions, inheritance, arrays and regular expressions while flagging traps such as a model built on global variables. Auston Bunsen recommends it.

Zed A. Shaw’s beginner course in Ruby runs through 52 exercises, each typed out by hand and debugged by the reader, building up from variables and loops to objects, automated testing and simple web programs. Auston Bunsen recommends it.

Miran Lipovača teaches Haskell, a demanding functional language, through drawings and jokes, starting with syntax, recursion and types and climbing to monads, applicative functors and zippers. Santiago Basulto enjoys it for the concepts, though Haskell is not part of his daily work.

Justin Richer and Antonio Sanso teach OAuth 2 by having readers build a client, an authorization server and a protected resource, then cover tokens, OpenID Connect, common vulnerabilities and securing REST APIs, assuming basic programming plus HTTP and JSON. Louis Nyffenegger recommends it.

Luke Welling and Laura Thomson teach database backed web development with PHP and MySQL, ending with full projects such as a shopping cart and user authentication. Scott Johnson credits it with teaching him the craft when a software engineering job arrived after the dot com crash.

Martin Odersky, who designed Scala, teaches the language one stage at a time with Lex Spoon and Bill Venners, moving from its basic elements to functional programming on the Java Virtual Machine and advanced features. Santiago Basulto values the concepts behind it even without daily Scala work.

Jon Bentley builds each essay around a small, nagging problem from real programming work and solves it with care for design and machine resources. James Stanley found many moments of sudden insight in its well-defined problems and efficient solutions.

Jean-Philippe Aumasson explains modern encryption in technical detail, from hash functions, block ciphers and randomness to RSA, elliptic curves and TLS, and every chapter shows common implementation errors with code samples. Louis Nyffenegger recommends it.

Athanasios Papoulis and S. Unnikrishna Pillai’s graduate textbook develops probability theory and random processes deductively, with applications for electrical engineers. Nassim Nicholas Taleb suggests it to nonmathematicians starting in probability and uses it as a desk reference.

Trevor Hastie, Robert Tibshirani and Jerome Friedman set machine learning inside a statistical framework, covering supervised and unsupervised methods from boosting and support vector machines to random forests and the lasso. Nassim Nicholas Taleb uses it as a reference and keeps two copies.

Frederick P. Brooks Jr. covers the management side of technical work, drawing on IBM’s System/360 to explain why adding programmers to a late project slows it, plus the second-system effect and prototyping. Jeff Bezos has recommended it.

J.E. Gordon introduces structural engineering for general readers, explaining in plain, witty prose why buildings, bridges and dams keep standing. It is a primer rather than a manual, and Elon Musk calls it very good for learning structural design.

With stick-figure drawings and only the thousand most common words, Randall Munroe shows how 54 machines and natural systems work, from microwave ovens to helicopters, an entry point to technical subjects rather than a manual. Bill Gates hopes its drawings send readers off to learn more.

Four security and agile practitioners, Laura Bell among them, focus mainly on process: fitting security into agile and continuous delivery, with security testing in every release and compliance in DevOps teams. Louis Nyffenegger recommends it.

Henrik Kniberg describes how a Swedish company of about 40 people adopted Scrum and mixed in XP habits like pair programming and test-driven development, with tips on backlogs, estimation, demos and testing. Mircea Scarlatescu recommends it.

Mike Cohn’s guide is about software process more than code, showing teams how to adopt Scrum and keep it working, from team structure and scaling to metrics and distributed teams. Nicolae Andronic recommends it for building products in an organized way.

Avinash Kaushik’s follow-up is a framework for web analytics, mainly aimed at making organizations data driven, with practical advice on analytical techniques, measuring social and multichannel campaigns, and running experiments. Raluca Radu recommends it alongside his earlier book.

Avinash Kaushik’s book is mainly about web analytics strategy for marketers, but its week-by-week program teaches hands-on skills, from core metrics and internal search reports to paid search, email measurement and testing, plus choosing tools and tags. Raluca Radu recommends it.

Brian W. Kernighan writes for readers with no technical background, explaining how hardware, software, networks and the web work, what programming is, and the privacy and security risks involved. Eric Schmidt calls it the clearest, simplest explanation of the digital world.