I worked with the data science association on their new standards for "Data Scientist" interviews (entry salary of $125,000), and both real questions and after interview polls were included for the biggest names in data today, from the web to corporate and government IT. "Data Scientist" is one of the hottest new jobs out there today, and some companies are even forming CDSO jobs--Chief Data Science Officer!
To begin, ALL FOUR of the books in this review are 5 star "superstars" for IT interviews. The two problems are, my library customers want to know the top two, and our Amazon shoppers want to know if they can get away with one, two, three, or if they have to buy all four! Of course the answer depends both on the focus of your resume, and the overlap/focus in the four books.
First, the summary, by author, title/Amazon link, year published/edition, number of pages, trim and cost, problems included, main language(s) foci. These four are the most frequently purchased by the over 100,000 libraries (including corporate technical libraries and schools as well as private and public) in our database. (Note: page counts are via visual inspection at the time of this writing, not Amazon stats. Pages can vary with on-demand books.).
Aziz, Elements of Programming Interviews: 300 Questions and Solutions by Aziz, Adnan, Prakash, Amit, Lee, Tsung-Hsien 1st (first) Edition (10/11/2012), 2012, 481 pages, 6 x 9, $25, 300 problems (mostly C++, concurrency in Java, discrete math in formulas and English)
McDowell, Cracking the Coding Interview: 150 Programming Questions and Solutions, 2011 (5th edition), 500 pages, 6 x 9, $23, 150 problems, (mostly all Java except of course the C, C++ question sections!)
Mongan, Programming Interviews Exposed: Secrets to Landing Your Next Job, 2013 (ed. 3), 301 pages, 7.4 x 9, $18, 150+ problems (C, C++, C#, Java)
All four of these fine prep texts cover the usual suspects in Algorithms and Data structures, including a focus on "scalable" problems of most concern to the Amazons, Googles, Facebooks, etc. of the world. These include recursion, arrays, lists, hash tables, binary searches and trees, and other foundation coding subjects.
All also cover the usual tricks, brain teasers, presentation problems, prep, process, etc. issues, and in the case of Cracking, specifics on many different company processes.
The divergence is in the "extras." Aziz jumps into parallel computing and covers discrete math (in grad school joke terms, all the computer oriented math that has been taken out of high school courses). McDowell has an unusually well written probability section. Guiness is very up to date with cross platform apps and concurrent programming nightmares, and goes into both more depth and detail on individual topics like big O notation. Mongan is published by wrox, and has not only technical editors, but outstanding web resources. His database section is the most robust of the group.
Aziz and McDowell are print on demand, which usually means there are many more errors in early going, but much faster correction of them via almost weekly files to the printer. Guiness is Wiley and bulletproof. We've tested the code extensively in all four (my payroy sister programmers, not me!) and ALL of them are outstanding, with very few errors at this writing, which can only get better fast in the two PODs, and wasn't a problem to begin with via the technically edited wrox and wiley teams.
Surprisingly, there is NOT a lot of overlap in solutions in these four texts, just as there IS a lot of overlap in the questions (strings, arrays, binaries, hashes... structures are structures and algos are algos). The difference in ALL these books (as opposed to a Cormen) is that the algorithm examples are not academic--they give you many options to "cheat" - and most of the cheats are more real world than techniques given in the 1,300 page algo function texts.
McDowell is the industry standard, but she teaches very much to Google, as does Aziz, meaning web focus, and even a little forgiveness on php, but NO forgiveness on memory or scalability. If you're a library client and have to pick two, we advise one from the McDowell/ Aziz dyad and one from the Guiness/ Mongan dyad. If you're applying for a job with a specific language requirement, these self sort, although of course all are object oriented today.
For shoppers preparing for a real interview: buy all four. I mean, come on. This is your future! You can get all four for the price you'd pay for a larger (way less useful) algo + data structure or individual language text, and maybe less. Some points about interview technique are common, but all four offer different and important examples in approaches to solutions, even though they share common algorithmic and data structure challenges.
IRONY: The only programming area growing faster than data scientist today is at the other end of the big scale spectrum: embedded systems. I kid you not, specialize in embedded, and you're GUARANTEED a dream job, both due to the explosion of these systems, and the rarity of programmers here (but yes, you have to get into circuits!). Our sister Payroy group shows job stats, demand and salaries that are to die for if you go there-- way better than Google. NONE of these books cover it (because other than mobile and server embeds, embedded was traditionally automotive and industrial, but even "Google and Microsoft TV" type ventures are now hungering for it).
There is NO good interview book out on embedded yet, but these two are the best of breed in the field itself: 1. Samek (Practical UML Statecharts in C/C++: Event-Driven Programming for Embedded Systems) and 2. White (Making Embedded Systems: Design Patterns for Great Software). Why C and C++? Because that's where the majority of electronics still reside, and "object" programmers in the field often just use the C subset of ++ and don't really get into sexy classes/methods/parents/kids, etc.! 6 months brushing up on this, specializing, and going for an embedded job will be worth years of competing with the interviews in these texts!!!
Now, a simple tip. I was part of a team that interviewed for a high level, very high paying digital art programming position at shader joes dot com. One candidate stood out as really technically challenged--she even confused a call with a register in one of her answers! She called herself an "autodidact" - meaning, unlike Yahoo, we can't be recruiting only from the 18 top schools.
At the end of her interview, she asked us to check out a disc she'd brought. She had programmed her own video game with movie-real characters, explosions, storyline, etc. using Unity, Maya, blastcode, Python, Lua and C#, with web distributions in Java, HTML 5 and php. She proceeded to explain her entire process, from idea to distribution. She was hired before she could reach the elevator. In olden-days, old timer parlance, don't forget your "portfolio" if you have one! It can trump a LOT of the bureaucratic hurdles!
EMAILERS ANSWER: IF you are a manager, rusty at coding, a data scientist, etc. and are in an interview where you have to "understand" coding basics, but not necessarily code, see our review of Karumanchi (Coding Interview Questions).
Library Picks reviews only for the benefit of Amazon shoppers and has nothing to do with Amazon, the authors, manufacturers or publishers of the items we review. We always buy the items we review for the sake of objectivity, and although we search for gems, are not shy about trashing an item if it's a waste of time or money for Amazon shoppers. If the reviewer identifies herself, her job or her field, it is only as a point of reference to help you gauge the background and any biases.