Introduction

Many of the readers here on Medium are looking to be a data scientist or data analyst, and are therefore interested in the interview process for each position. In my experience, I have interviewed with several companies for both roles. Below, I will detail the process for both roles and highlight where they are similar and where they are different.

Data Scientist Interview

Data scientists can expect to automate manual processes at their company. Using machine learning packages from libraries like sklearn and TensorFlow, a data scientist will ingest data, clean it, train and evaluate a model, and output suggestions or predictions for an end-user. Some of the process focuses on coding and algorithms, while the other part of the process focuses on soft skills like developing the business problem and explaining the results to an end-user. You will work with tools like Jupyter Notebook and programming languages like Python and R (sometimes SAS).

Overall, the process includes phone screens by the recruiter and hiring manager. Then, meeting with the team for a conceptual and coding interview (some interviews do not include coding, but will instead, require a take-home evaluation of a common business problem that you would be likely to encounter at the company in the future). Next, you will explain your findings to either another data scientist or an upper leadership member like a senior product manager. Lastly, you will summarize all that you learned from the previous interviews, as well as compile reasons as to why you are the best fit for the role.

Data Analyst Interview

A data analyst can expect to query database tables, perform joins, subqueries, and report on data requests. You will work with several stakeholders to gather requirements for each request. You will work with tools like SQL and PostgreSQL or other querying platforms, along with Tableau, PowerBI, and other dashboard tools.

The data analyst interview will share quite a bit of similar steps, especially in the process overall. However, the differences and uniqueness lie in the concepts and coding challenges of the interview. You will expect to have phone screens with your recruiter and hiring manager. Then, you will meet with other data analysts on the team to go over key concepts like databases. After that, you will perform a coding challenge, usually entailing common SQL questions that will require you to join and sub-query. Next, you will explain results to another data analyst or a stakeholder like a customer success manager. Lastly, you will discuss your previous interviews and what you thought about the team to the hiring manager, as well as outline the reasons as to why you should be a data analyst at that company.

Similarities and Differences

Process

As you will see below, both data science and data analytics interviews share a similar process. While the process may be similar, the details of the process may differ, which I will refer to later. Here are the commonalities of the steps in the interview process:

Commonalities

• initial recruiter phone screen — a call with the recruiter
• hiring manager screen— a call with your future hiring manager
• specialist concept interview — specialist from each role (another data scientist and data analyst from the team you will be working on)
• specialist coding interview — specialist from each role (another data scientist and data analyst from the team you will be working on)
• leadership interview — explaining results to non-technical users
• hiring manager final interview — wrapping up why you think you are a good fit for the role

common tools: SQL, Tableau, and Jira (ticketing platform)

Differences

While the process is the same, the main differences lie within the specialist part of the process, usually.

—specialist concept interview

Data Science: 

Common machine learning algorithms like random forest and logistic regression for example
Common buckets of machine learning and data science like unsupervised vs supervised learning
Data Analyst:
Differences between joins like inner, outer, left, and right joins
Sub-queries
Indexing
Group by
Where clauses

— specialist coding interview

Data Science:

Programming in Python (for loops, arrays, and functions)
Programming in R
Programming and commenting code in a Jupyter Notebook
Sometimes, instead of a coding interview, you will have a take-home assignment that will require merging of data, exploratory data analysis and cleaning, model building, and output and explanation of results
Data Analyst:
SQL quering
Tableau (not coding, but a specific skill where you will need to know common functions and calculations in the popular interface)

— leadership interview

Data Science:

Explaining models and its predictions to non-technical users
Data Analyst:
Explaining a query and its results to non-technical users

#machine-learning #data-analysis #data-science #interview-questions