Day-in-the-Life of a Data Scientist at Amazon

Sakshi GuptaSakshi Gupta | 5 minute read | July 8, 2020
Day-in-the-Life of a Data Scientist at Amazon

In this article

Amazon may have started out as an online bookseller, but today its business spans dozens of data-generating and collecting categories, each requiring teams of data scientists to make sense of how the company is performing and where it can go next. Read on to learn more about what it’s like to be a data scientist at Amazon.

Amazon is one of the most data-driven companies in the world, with big data informing the growth and strategy of all its businesses, from its online storefront, Kindle, and its home assistant Alexa, to Amazon Web Services, AmazonFresh, and Amazon Studios.

Because of this, data scientists play a pivotal role in the business—they use statistical methods to analyze customer behavior data and forecast business and shopping trends; they use artificial intelligence to help Amazon Web Services customers build their own machine learning models; and they write algorithms to teach Alexa to understand human questions and commands.

Springboard alumnus Aashish Jain joined Amazon as a data scientist after completing a data science bootcamp and quickly rose through the ranks to become a research scientist, focusing on Amazon’s Alexa.

Below, he shares his experience and offers advice that can help future Amazon data scientists succeed in the role.

What does a day in the life of a data scientist at Amazon look like?

Aashish Jain had a background in chemical engineering before he took a data science course with Springboard and landed a job at Amazon. After a brief stint as a data scientist—a role that he characterizes as focusing on analyzing data for insights and using existing machine learning models to make sense of information—he advanced to being a research scientist, which requires him to develop those models himself.

“I really like AI in general and doing predictive modeling,” said Jain, who is part of the team that works on Amazon’s Alexa. “If you know Alexa, you know that it’s a voice-based service, so it’s heavily based on AI. Without machine learning, you can’t do anything with it. I was very interested in machine learning, and that’s what drew me to it.”

A typical day for Jain starts at 9am and ends at 5pm, although he says that Amazon offers flexible work hours, so it’s not unusual for his colleagues to start work at 7:30am and leave at 3:30pm, or to start much later in the day. The structure of each day can vary, although, in an average week, there are sprint planning meetings to assign tasks and determine project deadlines; team meetings where team members can share what they’re working on and review each other’s progress; and weekly retrospective meetings to reflect on what has been accomplished, what could be done better, and what is left to do.

When he’s not in a meeting, Jain builds different machine learning models—a process that involves reading research papers, identifying which algorithms work best for the problem he’s trying to solve, and implementing them using Python.“It’s not just model building,” he said. “There’s data analysis, data cleaning—it’s the whole data science pipeline.”

Get To Know Other Data Science Students

Jonathan King

Jonathan King

Sr. Healthcare Analyst at IBM

Read Story

Esme Gaisford

Esme Gaisford

Senior Quantitative Data Analyst at Pandora

Read Story

Aaron Pujanandez

Aaron Pujanandez

Dir. Of Data Science And Analytics at Deep Labs

Read Story

Responsibilities and impact of data scientists at Amazon

Amazon has fourteen leadership principles that include having a sense of ownership in the company and its projects, always learning and being curious, insisting on the highest standards, delivering results, and having the backbone to disagree. According to Jain, these principles inform how data scientists and research scientists at Amazon work.

“With something like ‘ownership,’ it means taking the initiative and not relying on others to come up with ideas for you,” Jain said. “With a principle like ‘dive deep’, it means we don’t look at things superficially and we really care about attention to detail.”

One of the projects that Jain has worked on is on helping Amazon’s artificial intelligence, Alexa, with automatic speech recognition. Although Alexa is a voice-based service, Jain said that when a person directs a verbal order or question at Alexa, the AI needs to first convert the audio into text, and then apply natural language processing to the text to understand what is actually being said.“That was the part I worked on,” Jain said. “I was given the text and my job was to help Alexa understand what the user is asking for.”He also takes Amazon’s principle of always being curious to heart. Jain told Springboard that one of the ways in which his work has an impact on the company is through not waiting around to be assigned a problem to solve. True to taking ownership and valuing curiosity, Jain allows his own curiosity to lead the way.

For example, for any data science problem, Jain said there are two approaches:

  1. You’re given data and have to figure out how to best use it; or
  2. You identify a problem and think of what kind of data you might need to solve the problem

“You need to be creative,” Jain said. “You need to ask, I have the data, what can I do with it? Or what data do I need? Thinking about problems and what challenges you want to solve is really helpful.“When you join a company, you’ll initially be given some problems to solve. But, as you progress, you have to come up with your own ideas, define your own problems, develop your own projects.”

Tips for succeeding as a data scientist at Amazon

In addition to always being curious and taking the initiative, Jain recommends completing data science projects from beginning to end in order to gain experience with the entire data science pipeline.

Prior to joining Amazon, Jain worked on lots of different projects in an attempt to learn and master a variety of data science skills and concepts. Once he landed at Amazon, though, Jain realized it would have been more valuable to have just completed one larger project from start to finish.

“It could be your personal project or the Springboard data science capstone project, which helped me a lot in the interview process,” Jain said. “You don’t have to do multiple things. Even if you do just one thing, but you do it end-to-end, starting from getting your data, defining your problem, data cleaning, developing a model, evaluating that model, analyzing—everything—if you do it well, I think that boosts your confidence and shows that you know the data science pipeline very well.”

Since you’re here…
Thinking about a career in data science? Enroll in our Data Science Bootcamp, and we’ll get you hired in 6 months. If you’re just getting started, take a peek at our foundational Data Science Course, and don’t forget to peep our student reviews. The data’s on our side.

Sakshi Gupta

About Sakshi Gupta

Sakshi is a Senior Associate Editor at Springboard. She is a technology enthusiast who loves to read and write about emerging tech. She is a content marketer and has experience working in the Indian and US markets.