Think Twice Before You Accept That Fancy Data Science Job

The junior data scientist mentions her interest in neural networks and drops the term “heteroskedasticity” and (BOOM) she has the job. And now shes expected to produce results from a data environment that isnt quite data-science-ready.Unfortunately, there isnt much literature out there to help junior data scientists distinguish ill-fitting roles from roles that will help them grow. Im hoping this post help make this distinction.   The following are some characteristics you should run away from if youre looking to land a data science/machine learning role as a junior job seeker. Most of the following may be attractive to senior data scientists who may be up for the challenge, but may still apply to them as well.1) No data architectureIf the firm doesnt have an established data environment, then guess what? Nine times out of ten, youll be responsible for building this, instead of working on actual data science projects. This is one thing I see over and over. RUN.2) No defined objectivesIf, during the interview, they say something like “we dont really have actual projects yet, but we just KNOW that theres value in our data.” This is an indicator that they havent thought much about data science projects and thus, dont really have a plan for you to provide value. RUN.3) No data engineers or machine learning engineersSenior data scientists may be up for this challenge, but expecting a junior data scientist to be able to perform well in this environment is unrealistic. Having no data engineers or no machine learning engineers means youll be building data pipelines and focusing on writing SQL queries or moving data around. And when you do finish a data science project, since the firm doesnt have any machine learning engineers, youll be expected to put this project into production, which can be overwhelming unless you have prior software engineering experience. RUN.4) No comprehensive vision for the teamThis one can probably be lumped in with #2, but it goes further than merely not having defined objectives. This one speaks more to the data science manager. Ive been lucky to have great managers in my career, ones that defined a comprehensive vision for the team in terms of projects and personal growth, both of which are important. If the hiring manager doesnt have a clear vision for the team, then, junior data scientists, beware. You can spot this when the hiring manager says something along the lines of “data science is so new and were all just figuring it out” or “were still trying to prove the value of data science”.The other side of this is growth. You want to make sure your manager cares about your long-term growth. Ask if they have one-on-ones with their team. Ask about their long-term team goals. Ask about long-term projects. Ask about a mentorship program. Usually the answers to these questions will give you an idea for the vision of the team and whether or not youll grow to your potential in this role. Your long-term growth is just as important as what you work on in your day-to-day. If the hiring manager doesnt have a clear, comprehensive vision for the team, RUN.5) Focus on tools over problemsThis is a tricky one, because everyone wants to use the fancy tools and solve the biggest problems, so when a hiring manager mentions these, were instantly attracted. But if the team and manager focus on tools over problems, then this may signal that they are more concerned with building cool stuff instead of providing practical value. In my experience, managers that focus on problems over tools will ensure that data science projects are providing value to business owners. Let me go ahead and break something to you early on: no one outside of your team cares what tools youre using; they just want the job done. And a focus on tools over problems is a sign that the team isnt providing value to the business. Usually, if youre working on a mature data science team, then youre likely already using the cool new tools to solve business problems. I dont think that this alone is a deal breaker, but if its combined with any of the above, then RUN.6) Youre the first DS hire (with no plans to hire a senior)When I graduated and got my first job, I worked directly with a senior software engineer. Every single line of code I wrote crossed his eyes and although this was frustrating at times, every minute was a learning experience. I felt like I was maturing in dog years. Every month I learned more than I did in the prior five months. I was a sponge. If I have one piece of advice to anyone starting their career, it would be to find a job where there are senior level employees that you can learn from. This should be one of your main priorities as a budding data scientist and if the hiring team doesnt have plans to hire at least one senior data scientist then RUN.   1) Find your nicheFind a role where you can carve out an area of expertise for yourself. For example, if you want to specialize in natural language processing (NLP), then instead of working where everyone is an expert at NLP, consider finding a role where no one is an expert—youll stand out a lot more, and probably provide more value. But be sure that they have the data and business problems to support the area you want to specialize in.2) Interview themThis is where a lot of young candidates go wrong..They dont ask enough questions..I see it all the time..Junior candidates just dont know that theyre supposed to ask questions, but this is one of the most important steps in finding the right fit..Ask about the team..Ask about the company..Ask about the possibilities to be mentored or mentor others (equally important)..Ask about inter-department collaboration..Ask about the types of data science projects youll be working on..Ask about the successes of the team..Ask about the failures (diplomatically, of course). Ask about the vision of the team. Anything youre interested in, ask about. An interview isnt just a time where they ask you questions; its a time for both parties to ask discerning questions to make sure both parties feel its a good fit.3) Ask about the role specificallyThis section is such an important sub-section of the previous one, that I decided to dedicate an entire section to it..Its the most important thing I focus on when I go on interviews.. More details

Leave a Reply