How To Hire Exceptional Analysts and Data Scientists - Issue 153
An interviewing guide for hiring managers on finding, interviewing, and hiring top talent for data science and analytical roles.
Welcome to the Data Analysis Journal, a weekly newsletter about data science and analytics.
If I had to look back at my entire analytics journey, at all the analyses I made, dashboards developed, or documentation written what I remain the most proud of is the people I hired. Whether for my own team or for my stakeholders and partners, most candidates whom I interviewed and hired went on to make me and my team proud.
I have a unique philosophy for interviewing, and it’s one that some don’t agree with. It has proven itself many times over, which is why I’ve continued to utilize it when interviewing potential employees even to this day. Today, I will be sharing that philosophy, along with my interview questions, guides, values, and caveats.
This publication is for hiring managers who are looking and interviewing data scientists, data analysts, product analysts, BI developers, and other analytical roles. If you are currently interviewing, this guide should also be helpful for you to see what qualities or key points hiring managers might be emphasizing.
Analytics responsibilities and required skills change depending on the company's stage, size, and structure. Below I’ll focus mostly on analytical roles across reporting, analytical thinking, experimentation, and ML modeling.
Where to find data scientists and analysts today
Speaking from my experience, many strong candidates I hired were either recommended by someone I know or applied directly to the position. The lowest quality candidates I had to interview came from different 3rd party recruiting agencies or from random LinkedIn invites and messages. For this reason (which is simply based on my personal experience only) I don’t work with any type of recruiting agencies that offer already vetted or tested candidates.
I believe the best places to find rockstar analysts today are:
My newsletter (obviously) is aimed to upskill analysts and data scientists of all flavors and levels, who are investing their time to grow and stay in the loop with the industry demands. Feel free to drop me a note. I am currently mentoring quite a few amazing stars who are open for work and their skills and level are above average.
Meetups and virtual events focused on database usage, BI development, best practices for handling data, lightning talks, etc. I ended up hiring over 10 people whom I met at such events.
Slack communities. I am “lucky” to be in 20+ Slack groups, and I think almost every group has a dedicated hiring room. I found many kick-ass women this way. Thanks to
, the author of the newsletter, for putting this list together - A list of favorite data Hangouts. I’ll add more groups there.LinkedIn groups can take some digging and searching, but LinkedIn makes it easy to spot the right-fit candidates, especially in topic-specific groups (many of which are private) or via comments to data-related posts. You can search for relevant groups by keyword, location, and size. Some of the groups to pick a few:
⭐ A shameless plug, I am one of the investors in interviewing.io - a mock coding interview platform. It’s a marketplace that offers anonymous mock interviews with engineers from top companies and the best-converting channel for hiring engineers for employers. They don’t support analytical and ML talent yet, but hopefully soon! Such mock interviews are proven to be the most effective to prepare for technical screening.
Interviewing guide for data analytical roles
🚫 No take-home challenges or assignments
I am strongly against any type of take-home assignment:
It’s unethical and disrespectful toward candidate time, as they likely will spend more time on it than they should because they want to nail it and stand out from hundreds of other applicants. For no pay.
It’s easy to cheat to solve it today. You know how.
If you can’t test the skill during the actual interview, you are doing it wrong.
It tells the candidate that you are lazy and don’t put much effort into interviewing.
For every candidate who receives a take-home assignment, I urge you to communicate with the recruiter or hiring manager that you are open to another round of tech screening (if needed) to demonstrate your skills rather than doing the home challenge. Many companies have become open to this. Let’s fight this broken system.
📉 LinkedIn + Portfolio is the best resume
This one can be controversial for many, but I personally don’t read CVs and resumes now. For me, a candidate's LinkedIn page and portfolio (which simply can be on their GitHub, Kaggle, blog, or public dashboards) speak louder to their skills, interests, and experience.
Keep reading with a 7-day free trial
Subscribe to Data Analysis Journal to keep reading this post and get 7 days of free access to the full post archives.