You may have solid knowledge of data science, but if you don’t have an effective resume you probably won’t get an interview. Sadly, most job seekers put almost no thought into this critical document. Most recruiters and hiring managers scan a resume for a mere six seconds before making a decision whether to continue reading or throw it away. Does your resume make a powerful case for granting you an interview under such strenuous conditions? If you are like most data scientists, it does not. Fear not, this article will give you the information you need to stand out from the crowd and convince the reader to interview you! First, it’s important to understand what a resume is supposed to do and what it is not. The only purpose of a resume is to get an interview. That’s it. It’s not supposed to convince someone to hire you. That’s what interviews are for. Too many data scientists load up their resumes with 4+ pages of skills and experiences. Don’t do this! Remember that the reader is going to want to skim your resume in six seconds. The longer your resume is, the less likely they are to want to read it all.
Strategy Before Tactics
A common mistake people make in creating a resume is to consider it as the first step in a job search. Writing a resume is a tactic. And tactics should always come after strategy. The first step in a job search is to ask yourself the tough questions about who you really are and why someone would want to hire you. Some example questions are:
- What are your strengths and weaknesses?
- What kind of job do you want? (e.g., customer-facing, research, product team)
- What field of application do you want to work in? (e.g., marketing, biotech)
- What is the story behind your resume?
- Why would someone want to hire you?
These are hard questions and coming up with good answers is more difficult than creating a resume. But the good news is that once you have done this hard work upfront, they will make it much easier to write an effective resume. Note the last two questions in the list above. The best resumes tell a story and let people understand you a little better without you explicitly stating that you are, for example, “passionate about data science” (This is a much-overused phrase, by the way. Don’t use it.) My resume, for example, hints at a story of someone who wants their work to make a difference. There’s also an obvious interest in marketing. Hiring is not nearly as logical as one might think! The reader wants to know a little about the person behind the skills. It’s important to give serious thought to why someone would want to hire you. This should be on the front of your mind during the entire job search process. And a good answer to this is not, for example, that you have coded up a convolutional neural network for image recognition. That’s fine and all, but what people really care about are things like (1) will you make their life easier, (2) will you bring new capabilities to the team, and (3) will you be able to communicate your analyses to others. Don’t actually write the answers to these on your resume! These are simply questions to get you in the right frame of mind.
Structuring Your Resume
Once you have done the hard upfront strategic work, it’s time to start writing your resume. Remembering that someone will probably only spend six seconds reading it, you’ll want to put your strongest selling points right up front and your weaknesses near the end. Don’t worry about following some order imposed by a template or suggested by some “guru”. This is your document and you get to craft it as you see fit. One of my biggest selling points is a Ph.D. from Caltech. Some while most resumes put Education at the end, I put mine as one of the first sections on my resume. Most resumes I see start with some sort of summary statement or Objective. If you’re going to lead with this, make sure that it’s very strong. This is where the work you put in earlier of creating a story behind your resume and determining why someone should hire you will help greatly. I’ve seen many weak, generic Objective statements or, worse yet, ones stating that they intend to use the job to “develop the skills that will allow me to succeed” in data science. People want to hire someone who’s ready to succeed now, not after developing skills on the job! Because I’m a mid-career data scientist, I used a full two paragraphs to emphasize my (supposed) wisdom about what it takes to be a successful data scientist and how I “…excel in the full lifecycle of data science solutions, from pre-sales to casting a business objective as a data science problem to presenting final results.” I wanted to differentiate myself from younger applicants who might not have the breadth of experience that I do. It should go without saying that if you have a wide range of skills and experience with a variety of technologies that it should appear high up on your resume. As of this writing, I don’t have experience with Big Data technologies (e.g., Hadoop, Spark) so I placed Technical Skills after Accomplishments. By the way, the order of sections on your resume helps tell that personal story we talked about earlier. By leading with Accomplishments, I wanted to emphasize that I’m someone who gets things done and is routinely praised by superiors. I also wanted to make it clear that I feel that my data science skills are not an end in themselves but merely the tools I use to get the job done. One word about what skills and technologies to list on your resume: don’t make it an exhaustive list of everything you can think of. I’ve seen resumes listing “Microsoft Office” as one of their technologies. Everyone knows Office (or can learn it quickly). This cheapens your entire resume and makes it seem like you are “stuffing it” to compensate for something. Also, give some solid thought to how much skill and experience you should have with something before listing it. I once interviewed someone who listed Python as one of their skills, but when I asked how much experience they had it turned out they’d be coding in it for a mere three months! That made their entire resume suspect in my mind. (Needless to say, they didn’t make it to the next round). Your Experience section is a critical one. Depending on how long you’ve been practicing data science, this might be extensive or might be very short. If it is short, then you will need to do some projects on your own that you can describe on your resume. It’s best if you do a project that is out of the ordinary and/or of special meaning to you. Stating that you analyzed the deaths of passengers aboard the Titanic won’t impress anyone since it’s well known this is a standard Kaggle problem. Likewise, analyzing the sentiment of tweets appears in so many online courses and books that this, too, is not worth putting down. Try to brainstorm a way of combining your interests and hobbies with your data science project. If you like to run, perhaps you can analyze the results of various races by scraping race results from the web. If you like to travel, perhaps you’d be interested in analyzing flight data to see which airline carriers have to most and/or longest delays. Choosing something you are legitimately interested in will not only make working on the project more enjoyable but your enthusiasm will carry through when you describe your work to a recruiter, hiring manger, or interviewer. If you do have on-the-job experience, it’s best to describe results you achieved rather than what actions you took. Yes, this is a great place to list what techniques and technologies you have used, but you want to make it clear to the reader that results are important to you. Giving concrete numbers, such as “reducing customer churn by 27%”, implies that you will pay for yourself with the results you generate for the company. This section is also one that you should consider tailoring to the specific company you send your resume to. We’ll talk more about customizing your resume below. It’s important to remember that your resume will probably be skimmed so you should only put the most important things in the document. What happens if you have other material you’d like readers to know? That’s when you can put hyperlinks to other content you’ve created on the web. Examples include your LinkedIn profile, website and/or blog, GitHub repository, Quora profile, and so forth. Your LinkedIn profile and blog are great places to put multimedia, such as videos or slide decks of presentations you’ve given. You can use GitHub for your code, but also as a website. I used to be active on Quora and have written several highly-upvoted answers that express my philosophy of data science, so I turned those into blog posts and put a link to my homepage on my resume. It’s tempting to use your resume to dump everything you’ve ever done, but this is a mistake. Ideally, you should restrict yourself to a single page. That may require you to condense several years of work so that it all fits on a page. Also, look for things you can drop (e.g., older jobs, skills everyone should be expected to have). A single-page resume is not a hard-and-fast rule, but it’s a useful guideline. I’ve seen data science resumes that stretch on for six pages! As someone who was used as a first-line resume “screener” in my last job, that does not impress me. Quite the opposite. My feeling is that if you cannot summarize your professional life story in a page or two, what are the chances you can summarize a complex data science analysis down to a few slides? The last thing I want is someone who can’t boil things down to their essence and expects me to do that for them. Now, I actually used a two-page format. I probably could have fit it onto a single page if I tried, but I wanted to use a lot of “whitespace” when designing my resume to give it a clean look. Which brings us to…
A Few Words About Design
It’s very common to think about impressing someone when creating a resume, but it’s even more important to think about the reader’s experience. Whoever is reading your resume has probably seen a lot of them and the last thing they are looking forward to is one that is indistinguishable from all the rest. If you give even just a little thought to design, you can stand out and make the reader’s task more enjoyable. By the way, incorporating good design elements into your resume has another strong benefit: it demonstrates that you know how to communicate well with the written word. While most data scientists realize that communication skills are important to have, most of the time the only indication that the applicant understands this is when they claim to be a “good communicator” (More on this terrible phrase later…). Crafting an attractive and well-written resume is more than some mere “trick” to impress someone; it demonstrates you understand the importance of communication more than any assertion on your part. What follows is certainly not meant to be anything more than a few helpful design tips. But given the fact that so many resumes have been created without any thought to design, a few tips are all you need to be memorable. I mentioned before that you should try to restrict your resume to a single page. If you find yourself decreasing the spacing between paragraphs and reducing your font size in order to fit everything on one page (and you cannot cut any more content), it’s time to think about spreading things out onto a second page. My resume was two pages because I used a two-column format where the left column was almost completely empty! Far from being “wasted space”, it made the document less intimidating to read and shortened the width of the paragraphs, which made them easier to skim. You don’t need to use a two-column format, but I encourage you to play around with so-called whitespace and see what a difference it makes in readability. You can use whitespace to indicate what elements go together. Related elements should be closer together than unrelated ones. Fonts are a topic that seems to bring up a lot of needless debate, in my opinion. Consider using two different typefaces: one for headings and one for your body text. A few combinations that work well for resumes are:
- Georgia with Tahoma
- Rockwell with Gill Sans
- Rockwell with Futura
- Museo Slab with Verdana
- Baskerville with Arial
There are certainly many more combinations that would work. All in all, typeface selection is not a big deal, but using something other than the default of your word processor or overused, common ones (e.g., Times New Roman, Helvetica) is yet another way you can make your resume stand out. If you are looking for a great typeface and are willing to spend a little money, I asked my designer friend, and author of the #18 Amazon Best Selling book Design for Hackers, David Kadavy what he would suggest for a resume. His recommendation was Adobe’s Minion typeface. And rather than pairing it with a complementary typeface, he suggested using italics, bold, and font sizing changes to distinguish the various elements on your resume. But remember, typeface choice is very low on your list of priorities when creating a resume so don’t stress over it. And don’t use a font size smaller than 11 point. If you start running out of space, find something to cut so your material fits on the page without resorting to space-saving tricks like this. Remember Kadavy’s suggestion of using different font sizes for different resume elements. An example is to make your headings a 14-point sans-serif font while the text directly underneath is an 11-point serif font. A few words on alignment. Good designs have strong invisible lines where elements line up with each other. A good example is the left margin. Resist the temptation to align the text on both the left and right edges. So-called justified text is hard to skim. And use center alignment sparingly, if at all. It can make your professional resume look like a high schooler’s report. Use typeface and font size changes to indicate headings, not centered text. Lastly, the best designs are ones that don’t draw attention to themselves. Some people encourage the creation of overly-creative resumes, like ones created to look like infographics. You’d have to be very careful that this would be appreciated by the recipient if you decided to do this. For a data science job, it generally won’t work well and will make you look too “artsy”. The only possible situation where it might – might! – work is for a company whose website states they hire unconventional thinkers. But I think creative resumes are generally a bad idea for data scientists.
Customizing Your Resume for the Job and Company
Here’s some advice that everyone knows they should do but few actually do. If you want the best chance of landing an interview, it’s best to take a few minutes and customize your resume for whatever position you are applying for. I recommend having several similar resumes on hand that you can quickly individualize. For example, you could have one resume designed for small startups, one for mid-size companies, and one for large ones. When you apply to a startup, you simply tweak your “startup resume” slightly and send it off. That’s much easier than having a single resume and having to do a major customization for each company you apply to. How exactly does one customize a resume for a company? Paying attention to the wording on the company’s website and job description is a good way of understanding what their ideal candidate is. If they talk about needing people who can handle many different responsibilities – like many startups do – then it makes sense to talk about how you like to take initiative. That trait may not be as desired at a large Fortune 500 firm that has a lot of rigid systems in place. Choose the same wording that they do when referring to skills and experience. You never know if the person who receives your resume is someone without a technical background who doesn’t realize that text analytics and natural language processing are close enough to grant you an interview.
Some Common Resume Mistakes
Although the emphasis of a data science resume is usually the technical skills and experience, soft skills and attributes are important as well. The problem is that while it’s natural to list technical skills on your resume, doing so with soft skills can seem odd. One thing I see a lot of is people referring to themselves as “passionate about data science.” Passion is a very good thing, of course, but it would be better to demonstrate this through your experience than to merely assert it. With soft skills, it’s usually a good idea to “show, don’t tell”. Think about how you can describe your past experience – whatever it may be – in a way that demonstrates your soft skills. Perhaps one of the most overused soft phrases on resumes is “great communicator”. Communication skills are essential for data scientists but this is such a generic statement that it’s meaningless. Are you skilled at giving presentations? Writing reports? Briefing non-technical superiors on technical material in a way they find interesting and understandable? Interacting with subject matter experts? I assert that anyone who is truly a “great communicator” would never use a statement as vague as this! Be specific about in what ways you consider yourself a great communicator. The incredible demand for data scientists has led to an explosion of massive open online courses (MOOCs) from sites such as Coursera, Udemy, DataCamp, and more. While these courses can be a useful part of your overall development as a data scientist, I recommend that you not put them down on your resume. It’s no secret that it’s pretty easy to get a certificate of completion for most of these courses so doing so is not the monumental achievement these platforms would have you believe. It’s best to simply state what data science skills you have without saying you got them from taking an online course. Lastly, listing the same skills as everyone else will make it difficult to stand out. Take the time to learn something unusual in data science. Examples include social network analysis, Bayesian networks, survival analysis, advanced ensemble methods (something more than random forest), and so on. Data scientists love to learn new things and positioning yourself as someone who can share new knowledge with your colleagues is one way to make yourself look unusually appealing.
Have Fun Creating Your Resume
Since much of the job search advice out there is a list of don’t-dos, it’s only natural that many people regard creating their resume as an arduous task. It doesn’t have to be this way. Remember, your resume is your opportunity to make a great first impression. If you follow the guidelines in this article, you’ll be well on your way to having a resume that makes you stand out from the crowd and piques the interest of the reader. Take pride in your resume and enjoy your job search! Your enthusiasm will be apparent to the recruiter or hiring manager, even if they are simply reading a document. I hope this article has helped you and please comment below with some of the things you’ve discovered made your resume stand out.