Google always keeps looking for talented people to hire them as data scientists. It's never easy for Google to find people with enough talent and passion. Let's look at the criteria to secure a job at Google as a data scientist.

As you may know, the interviews at Google are pretty intense. It sets the hiring bar high. Therefore, I've listed a few points before appearing for an interview at Google.

Know Your Statistics

An individual is expected to know Mathematics like linear algebra, and calculus more or less to get hired as a data scientist. They look for such people who live and breathe probability and statistics. Promising candidates require having the equivalent of at least 3 or 4 courses in probability and statistics, or machine learning. Anything more than these is the icing on the cake. One should be able to ace the homework and exams in his/her probability and statistics courses. Most of Google's data scientists already teach these courses before entering the eco-system of Google. There are a few websites available, such as stats.stackexchange.com, on which one can discover some questions and discussions to acquire his/her statistical skills.

Anything less than that could be augmented with courses in technical fields like economics, computer science, and engineering. Original research can also aid.

Get real Experience

One needs to demonstrate his/her work experience on real-world data. Although coming up with a new regression estimator for a few UCI datasets is impressive, those datasets are often used for comparing methods and not for getting real-world exposure. Google wants to see something that one had an opportunity to get his/her hands dirty on lots of real data. This indicates one has spent enough time to collecting own data after cleaning, sanity-checking and then making use of it.

One must write a script to pull data from one of Google's public API and make documentation about the discovery. One is suggested to use a web scraper to remove a few thousand web pages and fit a few topic models to create a news recommendation engine. One also can write an application for one's phone, which analyses it after tracking the usage.

Be Proficient at Coding

Even if a data scientist is not a hardcore engineer, Google makes sure he/she is capable of coding before getting hired. The best way to demonstrate the skill is to know to code ahead of time. For instance, one can point out the recruiters of Google to GitHub to showcase his/her coding skills. One is typically expected to become familiar with scripting languages such as Python and SQL. One or more numerical languages such as R, Julia, Matlab, or Mathematica are a plus. If someone knows a compiled language such as Java or C++, that's a bonus. If someone wants to enhance his/her coding skill, Khan Academy or other coding resources are highly recommended.

Be Passionate

Meeting the criteria would be easy if someone is passionate about issues related to data science. Probably one has spent a few years studying problems for which data provides a natural solution. Probably one has written code to interface with public APIs, from Google. Ideally, one has not only to be passionate about the methodology used to frame the problem, but also the problem itself.