Walmart Inc. is an American multinational retail corporation and the world's biggest retailer with over 20,000 stores and clubs in over 28 countries worldwide. Founded in 1962 and headquartered in Bentonville, Arkansas, the company operates a chain of hypermarkets, discount department stores, and grocery stores across the world along with Sam's Club retail warehouses. According to 2019 statistics, Walmart is the world's largest company by revenue, with US$514.405 billion, and also the world's largest private employer with over 2.2 million employees.

Walmart is also a data-driven company. With over 70 million different products sold at Walmart and over 275 million weekly customers, Walmart generates a huge amount of sales data across stores around the world daily. To make sense out of the data, Walmart launched the Data Café Analytics Hub, a state-of-the-art analytics hub that can process 2.5 petabytes of data every hour [1]. Within this data café are over 200 streams of internal and external data that can be manipulated and visualized to generate business-impact insights.

The Data Science Role at Walmart

Walmart has a data science and analytics departments with many roles like data scientist, data analyst, big data engineer, and tech architect embedded within their different product teams. The technology teams fall under an umbrella named Walmart Labs with teams such as:

  • Customer Technology
  • Merchant Technology
  • Supply Chain Technology
  • Business Engagement & Strategy
  • Global Cloud
  • Global Data and Analytics Platform

General data scientist roles at Walmart span across business analytics, statistical modeling, big data analytics, machine learning, and deep learning implementation with skillsets and tools ranging from simple data analytics/business intelligence tools to machine learning implementations.

Required Skills

Although the requirement for hire at Walmart may to some extent depend on the needs of a particular team, Walmart usually has a general hiring requirement for a data scientist position. The minimum requirement for a data science position at Walmart is an undergraduate degree with at least three years of experience.

Other relevant requirements include:

  • 4 years' experience with SQL and relational databases (for example, DB2, Oracle, SQL Server, etc.).
  • 4 years' experience with statistical and Numerical programming languages (for example, SAS, R, Python, and SQL)
  • Ability to work with large data sets, scale algorithm to a large dataset, and a sound understanding of big data technology stack.
  • Excellent working knowledge of statistics, mathematics, and machine learning algorithms.
  • Understanding of cloud computing platforms and large-scale databases
  • Experience in automation and scripting techniques
  • Proven background in Distributed Computing, Data Warehousing, ETL development, and large scale data processing

The Interview Process

Image from Walmart Lab's Medium

Walmart has a similar interview process like most big tech companies. Generally, the interview process starts with a 30 minutes initial phone call with recruiting, and then a take home challenge or technical interview with a team manager. After passing the take home and technical interview, you will then progress to the onsite interview comprising of 4 to 5 one-on-one interviews with various members of the team.

Initial Screen

Walmart initial phone screen is done over a 30 minutes phone call with a recruiter. Most often, this interview is non-technical in nature and may include a brief run-down of your resume, discussion of past projects relevant to the team you are applying for, and potential team members you will be meeting with. Note also, that there may be some basic technical elements in the interview to make sure your resume matches your experience.

Technical Screen and Take-Home Challenge

Generally depending on seniority of the role, Walmart may send a take-home challenge after the recruiter call to make sure you can pass a technical baseline. This challenge is most likely done in Hackerrank and will involve writing Pandas and Python code to manipulate a dataset.

Want to practice take home challenges? Check out Interview Query's take-home challenges.

Walmart’s technical screen is done via Hackerrank as well with a data scientist over video chat. The interview is usually one-hour long, and questions involved around this interview include data structures and algorithms question as well as SQL and possibly a discussion about machine learning at the end.

Prepare by practicing coding in a shared environment and thinking out loud about how you approach solving each problem.

Examples Questions:

  • Print all the branches in a binary tree.
  • What is multi-collinearity, how do you fix it in a regression model?
  • Explain the difference between bagged and boosting models?
  • What is the relationship between sample size and margin of error?

Onsite Interview

The onsite interview is the last interview in Walmart's data science hiring process. This interview comprises of 4 to 5 interview rounds with data scientists and a hiring manager, all lasting approximately 45 minutes with a lunch break in-between. Interview questions in these rounds revolve around probability and statistics, SQL and data analysis, machine learning concepts, product sense, and general culture fit. Most times however, the onsite interviews rounds are tailored-specific to meet the needs of the team.

However, in general Walmart's onsite data scientist interview looks like this:

  • A technical interview with a team member involving a case study SQL-based questions.
  • A probability and statistics round with questions around regression analysis.
  • A technical interview round with a hiring manager, which involves answering questions on the fundamentals of machine learning algorithms and design (e.g. how would you design a recommendation system?)
  • An interview with a product manager on past projects relevant to the team you are interviewing for along with culture, values, and product metrics.

Notes and Tips

Remember, the interview aims to assess how you can apply various data science concepts to the company. In this case for Walmart, their biggest problems are in improving service delivery and customers shopping experiences.

Walmart tends to be a large company that allows for a lot of flexibility in different data science roles. Because Walmart has such a large multi-conglomerate business, you can find many opportunities to move around or move up in career ladder.

Lastly while Walmart's offices are headquartered in Arkansas, Walmart Lab is headquartered in San Bruno, just north of the heart of Silicon Valley. Walmart's offers are still competitive with the greater landscape as they need to attract talent. Salary packages can be negotiated with RSUs + high base compensation.

Walmart Data Science Interview Questions

  • Tell me the difference between supervised and unsupervised learning.
  • Describe what the DBSCAN algorithm does?
  • What is the ROC curve? What does a confusion matrix do for model evaluation?
  • What is the difference between gradient boosting and random forest?
  • A person using a search engine needs to find something. How do you come up with an algorithm that will predict what the user needs after they type only a few letters?
  • What are the general assumptions of a linear regression?

References

[1] Really Big Data At Walmart