Cracking the Analytics Engineering Interview
How to break into the hottest data job of 2023 🔥
As the modern data stack becomes more popular, more organizations find a growing need to hire an Analytics Engineer that can be a liaison between business and technical teams, help them build out and maintain their data transformation layer, and overall have a solid understanding of how data interfaces with the business.
I’ve spent the last few months interviewing for the Analytics Engineer role across different organizations and wanted to put together an interview guide for folks considering transitioning to this role or actively interviewing. While this is purely based on my interview experiences, the process is more or less the same across all companies hiring for this role. This guide includes the kind of questions to expect, things I did well, and some mistakes I made that you can avoid.
Quick context on my background—I started my professional career as a Software Engineer and accidentally found myself in the world of data and analytics while working on some ad-hoc dashboarding requests. I knew I wanted to transition to a role at the intersection of business and engineering and that’s how I ended up as an Analytics Engineer.
Overview
The typical interview process for an Analytics Engineer involves 3–5 stages (a combination of the following):
Recruiter call
Hiring Manager
Technical interview/ take-home assignment
Stakeholder interview(s)
Recruiter Call
This is an initial 30 - 45 minute call with the recruiter to understand your background/ experience and share more about the role and the company. It is usually an informal conversation where the recruiter wants to understand your motivation for applying to the role. The call will typically follow the pattern below:
Information about the company and the role [10 mins]
Candidate introduction/ background [5 mins]
Specific questions related to experience/ motivation [10 mins]
Time to ask questions [5 mins]
The last few minutes are a good time for you to ask any high-level questions about the company that you’re curious about/ could be dealbreakers (funding/ runway, visa sponsorship, remote work, PTO, growth opportunities, etc.)
Do:
Show your enthusiasm for the company and the role.
Prepare a 1–2 min long pitch about yourself and your experience relevant to the role (you’re going to have to repeat this for every single interview so make sure you have this nailed down).
Have 3–4 good questions ready for the interview. This is the time for you to assess if the company is a good fit for you.
Don’t:
Say negative things about your current or ex-employer regardless of the reason you’re looking for a new role.
Talk too much. Practice your elevator pitch and keep all your responses short (1–2 mins ideally).
Talk about compensation right away. You’re still trying to see if the role and company are a fit for you.
This is assuming that salary ranges are a part of the job description and you have a vague idea of whether or not your expectations fall within the range. If the job posting does not have a salary range, I strongly advise you to talk about compensation at this stage and ask the recruiter for a salary range. Don’t waste your time interviewing for a role where you are not compensated fairly for what you bring to the table.
Example Qs:
Tell me about yourself and why you’re interested in <company>
Why are you looking for a new role/ What are you looking for in your next role?
Do you have experience building ETL pipelines?
Call with Hiring Manager (HM)
In my opinion, this is probably one of the most important stages of the interview. This is your opportunity to get to know your future manager 1:1 and ask specific questions related to the team you’ll be working with. Come prepared with 3–4 thoughtful questions which will give you a good sense of the HM’s working and communication style, how the team operates, current challenges and focus areas, vision for the team, etc.
Remember: You’re interviewing your manager and team as much as they’re interviewing you.
For me, these calls have been technical conversations, mostly trying to understand my experience specific to Analytics Engineering. Expect to answer questions about your processes and workflows, experience working with different data tools and cross-functional stakeholders, etc.
Do:
Have some of your past projects ready to talk about (Use “I” instead of “we” - the interviewer wants to know what you did)
If the tech stack that the company uses is not mentioned in the job description, ask the hiring manager about this in detail. The use of older tools vs. modern data tools will give you a good insight into the team culture and the kind of work you’ll be doing.
Have an opinion about some of the past tools that you’ve worked with (what you liked about them, what you didn’t like about them, etc.)
One of my interviewers and I bonded over our shared dislike for a tool both of us had previously worked with.
Don’t:
Lie about your experience. If you haven’t worked with a particular tool, say you haven’t. It’s totally fine to say “I don’t know” in an interview and they won’t always hold it against you (unless you say you don’t know SQL and that's a different conversation 👀 )
Hot tip: If I can’t answer a question, I usually say “I don’t know the answer to this question but if this were a real-life situation, here’s how I would go about figuring it out…”. This shows your curiosity and eagerness to find a solution to a problem even though you may not have all the answers in hand.
Example Qs:
What is your experience with building self-serving tools for non-technical stakeholders?
What data stack have you worked with before?
What’s the largest size of data you’ve worked with in the past?
Have you worked with <insert data stack that the company uses> before?
Walk me through a project that you worked on in your last role.
Walk me through your process of building a dashboard.
How do you prioritize incoming requests from stakeholders especially when you and the team are resource constrained?
How do you deal with ambiguous requirements from stakeholders?
What is your experience with dbt? (PS: If you don’t know dbt, you must!)
Technical Screen / Takehome Assignment
This is a crucial step to test whether you are technically qualified to do the job or not. Depending on the company, this can either be a live-coding interview or a take-home assignment.
If the interview process does not include a technical round, this might be a red flag that the organization may not know what they are looking for and expectations can be misaligned in the future.
In a live-coding interview, you’ll be given a business objective and will be asked to design a data schema and write SQL queries to answer some business questions.
Do:
Practice, practice, practice! There’s no other way to get through this stage.
Ask clarification questions. Interviewers will purposely leave out specifics because they want you to ask questions before you start writing your SQL queries.
Read the question multiple times and think out loud after you’ve read it. I’m an internal thinker so this was one of the things I had to practice and be very conscious about.
Get to a working solution first and then optimize. Oftentimes, the ticking clock can freeze your thought process (I’ve been there!) so it’s better to have something for the interviewer to look at than let perfection get in the way and have nothing to show.
Consider your interviewer as a thought partner. If you have an idea of how you’ll solve the question, say it out loud to the interviewer first before diving into writing the SQL. Sometimes, they are more interested in understanding how you think about the problem than if you can write accurate SQL (this is necessary but you’ll also have access to documentation if you need it on the job)
Story time: In one of my live-coding interviews, I completely screwed up part 1 of the question because I was very nervous. I only had 15 minutes left until the end of the interview and I still had one more question to go. To save me some embarrassment (at this point, I knew I was not going to move to the next round because it took me 30 mins to figure out the first one) and not end the interview early, the interviewer asked me to think about the second question knowing very well that it was a lot more complicated than the first one and I just didn’t have enough time to answer it. Somehow, I managed to provide a solution in under a minute. He was very impressed since no other candidate had been able to crack that one so quickly. Fast forward, he moved me to the next round! (If you’re reading this, thank you KC!)
Don’t:
Start coding right away!
For a take-home assignment, you’ll be given a publicly accessible dataset and a few questions that will require you to write SQL queries against the dataset. For this exercise, organization, and documentation are as important as writing the correct SQL query.
Do:
Send clarifying questions back to the hiring manager once you’ve had the chance to read through the assignment a couple of times. If there isn’t space or time to do this, clearly state your assumptions with your solution.
I’d highly recommend creating a GitHub repository with your SQL queries. This will help you organize your data dictionaries, queries, output, and assumptions for each question.
Be “extra”. Make use of comments as much as you can. This ensures readability and shows that you empathize with your stakeholder. Take screenshots of the output if you’re using another tool to write queries and include them in your solution.
Package your solution. This exercise is not just about writing SQL queries, it’s about showing how good you are at Analytics Engineering. When I’ve interviewed candidates in the past, there has been a clear distinction between candidates who I want to move to the next stage vs. candidates I’m not particularly excited about just by looking at their approach to solving the take-home assignment. So, try to put in as much effort as possible. It shows.
Don’t:
Write bad/ inefficient code. Use CTEs, and comments where possible.
Indentation matters. Empathize with the person who will read through your solution and make it easy for them to follow along.
Stakeholder interview
This is usually the last stage of the interview process where you’ll meet with 3–5 different cross-functional stakeholders that you will closely work with. Think data team members, product managers, operations, engineering, revenue, etc. This interview involves high-level technical conversations and behavioral/ situational questions. Sometimes it can be an hour-long panel interview with 3–4 people on the call or it can be multiple 30-minute calls with each stakeholder.
Do:
Let your personality shine! Stakeholder interviews can be super fun and an opportunity for you to get to know the type of people you’d be working closely with.
Try to understand what information is most valuable to them. Everyone has different experiences and challenges with data. Ask questions to figure out how they use data and what they need help with.
Don’t:
Take this stage lightly. You’re still in the interview process and treat it like one.
Try to use fancy words to explain your past projects and processes. Analytics Engineering requires you to effectively communicate with both technical and business teams. Talk about your experience and projects in plain, simple language so that your interviewers can understand you.
Example Qs:
What kind of data have you worked with in the past?
How do you handle sensitive data?
What is your process for triaging a bug in a dashboard?
How would you explain a technical problem to a non-technical stakeholder?
Walk me through a project that you’ve worked on that required collaborating with multiple stakeholders.
What steps can you recommend to ensure high data quality?
Know before you go
SQL: I used Codewars and Leetcode to practice SQL. Make sure to brush up on intermediate SQL concepts like window functions.
dbt/ data modeling: You’ve got to know dbt well for the interview. Check out their online courses and certifications to familiarize yourself with the tool.
Python: This is not a requirement to thrive in an Analytics Engineering role IMO but a good skill to have anyway.
Past projects: Have 2-3 projects ready to talk about. You will definitely be asked to talk about your past experience working with data and building processes.
Communication: Collaboration and building trust with cross-functional stakeholders is a key part of being an Analytics Engineer which means your communication skills need to be top-notch.
Version control: Analytics Engineering takes a software engineering approach to analytics so knowing Git is crucial.
Modern data stack: You don’t need to know every single tool out there but you should be familiar with at least one tool in each category below.
Ingestion: Fivetran, Airbyte, Segment, Stitch, etc.
Transformation: dbt
Warehouse: Snowflake, AWS Redshift, Databricks, Google Big Query, etc.
BI tools: Sigma, Tableau, Looker, Mode, etc.
This is not an exhaustive list but just some modern data tools to think about as you evaluate the kind of company and team you want to join.
Additional Resources
https://www.getdbt.com/data-teams/hiring-analytics-engineer/
https://medium.com/geekculture/so-you-want-to-be-an-analytics-engineer-practice-this-3cf8f4829895
I hope this gives you some insight into the interview process for an Analytics Engineer role. If you have any questions, suggestions, or feedback, feel free to leave a comment here or reach out to me on Twitter or Linkedin :-)
Huge thanks to Chase Sommer, Alex Stenlake, Shreya Datar, and Dave Flynn for the feedback!