10 Things Data Science Job Seekers Should Know

You're probably rolling your eyes: "great, another top 10 list on LinkedIn". Trust me, I write a top 10 list only if I really believe the 10 points are valuable and necessary. I also write this now because there are so many people seeking data science positions both at the entry and more senior levels. This is not intended to be a "how-to" guide but should be read as a "things I wish someone had told me before I sent off 500 applications and failed a bunch of interviews" guide. The pandemic has produced an incredibly challenging situation. We all owe it to each other to contribute and support as we are able.

Prior to reading this list, I want to note that I am not the expert on job searching. What I am sharing here are general principles I've learned in my own job searching experiences and from listening to and speaking with others. None of these examples have an "N of 1", so to speak and are supported by many observations. That also means they don't apply to every situation, but I hope you find them useful.

1. There is no magic pill for a data science position. It takes persistence and an ability to learn from failure but not let it drag you down. We don't have the equivalent of "doing 300+ Leetcodes" that software engineering does to get you ready. Regardless of how much you prep, how much you think you nailed it and how sure you are of yourself, you WILL be frustrated along the way. Maybe you'll forget what a p-value is for a minute, or you'll later figure out you needed to use a self-join or like me, you go into an interview not expecting to see a linked-list question. It happens. How you respond to that frustration will determine how it goes.
2. Ideally, you know what type of position you want and why. I know it is tempting to "apply for everything" and it is true that the job search process is often a numbers game, but even with that, you should define what type of role you want and why you want it. You may not end up in the perfect position but you should think about how a role you enter into might lead toward that "dream" role. Maybe you want to be an ML engineer - is there a path from analyst to ML engineer? Are there examples of it being done at your target company?
3. Your resume is important but there is no "perfect" way to present it. At its core, the resume is a way to present your skills and your impact in positions you've held before. Ask yourself: does my resume show that I know the right tools, display the right technical leadership and demonstrate that I had a positive return on investment with my previous work? Try to do this without listing every skill under the sun (R, Python, SQL, +100 other skills that no one could master given 5 lifetimes). The font, the color, the formatting and the size of your name on the resume won't make or break you. The content will.
4. Data science is much bigger than Google, Facebook, Apple, Amazon and Netflix. Don't get me wrong, the data scientists that work there are often world class and it is an amazing experience to work alongside talented people. I had such an amazing experience at LinkedIn. However, not everyone can/will or should want to get into those companies. I've failed interviews for a couple of them before and found extraordinarily good matches elsewhere. If you judge your ability to work with data based on whether you passed interviews to get into these extraordinarily hard to crack companies, you'll be disappointed in yourself when you shouldn't. Don't aim for prestige, aim for a place that will allow you to really shine.
5. Be very conservative in how often you reach out to recruiters and hiring managers asking for them to review your application. Definitely do not send a message saying "Do you have any positions that fit my profile?" That will not go well and is not respectful to the recruiter or HM. If you do reach out, have a short ask. Use 3 sentences: who you are, what position you're interested in and an ask of the person. Don't start with "hello" or "are you there?" - it doesn't work. With a field like data science full of talented people it is normal to want to find a way to stand out. Just make sure if you do so it is done in a professional way.
6. Do you know what the position is and what it requires? There are so many weird titles in the data science space. These include variations of analyst, scientist, ml engineer, data engineer and any combination of the previous words that you can name. Here are some common expectations and skillsets evaluated for each role from a previous post: https://www.dhirubhai.net/posts/eric-weber-060397b7_datascience-machinelearning-jobs-activity-6653705485327446016-shpW
7. Are you applying for a specific role or is the interview process followed by a team matching phase? At many large companies the interview process occurs prior to your placement with a team being determined. They screen you for fit and skillset and only place you with a team once you've been approved for hire. Make sure to talk with your recruiter about this and how it occurs so you understand the steps of the process. This team matching can dramatically increase the time required to finalize and sign an offer.
8. Most interview processes require the following: technical phone screen, product phone screen, statistics, data manipulation (SQL, R, Python), algorithms, machine learning and sometimes company specific interviews. However, the content of these specific interviews is variable. There are hundreds of posts and lists out there with advice for how to prep for these interviews. Coding interview prep (https://www.dhirubhai.net/posts/eric-weber-060397b7_datascience-machinelearning-interviewing-activity-6661647124574416896-lxiv) and SQL prep (https://www.dhirubhai.net/posts/eric-weber-060397b7_datascience-analytics-sql-activity-6663082042990952449-wuOK)
9. Feedback and decisions post-interview take time. It is really hard to be patient. If you have deadlines or other offers, let the recruiter know but try not to send an email every 12 hours seeing if something changed. Right now, companies are being extremely careful with their budgets and know that they have candidates who are competing for a smaller number of positions. They will take their time. I understand patience in a pandemic is one of the hardest things to do, but if you expect it ahead of time it is easier to process.
10. Failure is a bitter pill to swallow but remember it is not personal. I used to take rejection for data science roles as a commentary on my skillset. I now understand that teams are doing the best they can to find a fit with the organization and that organization's mindset around data science. The hiring and decision process is noisy and many decisions end up being false-positives or false-negatives. If you take this personally and hold a grudge against a company that "didn't see your skills for what they were" you are wasting their time. I guarantee they are not spending anymore time worrying about you after the decision. It is important to move on, learn what you can from the process and continue to fight until things line up. They will. I promise.