data science bootcamp

Data Science Bootcamps: The Good, the Bad, and the Ugly

This is a guest article by Charlie Custer from Dataquest

It's been nearly a decade since the Harvard Business Review deemed data science “the sexiest job of the 21st century.” In the intervening years, not much has changed. Data science made the top ten in Indeed’s ranking of the best jobs in 2020, and broke the top three in Glassdoor’s 2020 job rankings.

From a career-switcher’s perspective, data science in 2020 offers more or less the same appeal that it did in 2012: It’s a highly lucrative, in-demand field that doesn’t require specialized degrees. In fact, it’s entirely possible to learn data science skills online through self-directed study.

Enrolling in a structured program is often perceived as being easier, though, so the boom of data science jobs has led to a corresponding boom in data science bootcamps.

Now, would-be data science learners are confronted with an overwhelming number of choices. So, what’s the deal with data science bootcamps? Let’s dig into the good, the bad, and perhaps even the downright ugly.

What is a data science bootcamp?

The term “bootcamp” is applied to a wide variety of data science learning programs. The “traditional” model, which is based on software engineering bootcamps, is a roughly three-month, in-person learning experience, during which students typically live with other students, studying full time, and receiving career mentorship.

These programs tend to be quite expensive; tuition is often $10,000 to $20,000. Since they require your presence in person, there are also food, travel, and lodging costs to consider.

Of course, there are also “online bootcamp” programs that attempt to replicate the three-month speed-learning experience in an online format, and hybrid programs that offer various combinations of online and in-person courses.

Online learning programs such as course sequences on Coursera or EdX and interactive platforms like Dataquest, Codecademy, and DataCamp are also sometimes referred to as “bootcamps,” although they’re functionally quite different. These programs tend to offer self-paced learning and fewer options for career assistance, although they typically also cost significantly less.

For the purposes of this article though, we’ll be looking at more traditional offline and hybrid online/offline bootcamps — those that aim to teach the required skills for data science in the space of a few short months and require a full-time or near-full-time commitment.

Will a bootcamp improve my resume?

The short answer is yes — although perhaps not in the way you think.

In general, employers are not likely to be wowed by the presence of a particular bootcamp on your resume. With no universal standard of bootcamp accreditation or student assessment and certification, there are simply too many different bootcamp programs out there for employers to draw conclusions about your skills based on the bootcamp name on your resume.

There are exceptions: Specific bootcamps may have relationships with specific employers that can help your resume stand out in a crowd. If a particular employer has hired someone from your bootcamp before and had a good experience, they may be more inclined to view your resume in a favorable light. Most likely though, adding a bootcamp or certification program’s name to your resume isn’t going to make a big difference.

Where a bootcamp will improve your resume is in the skills and projects sections. Employers are primarily interested in whether or not you can do the job they’re hiring for. That means that when they’re looking at a resume, they’re looking at:
  • Whether the skills you’ve listed match their needs
  • Whether the projects you’ve listed show you have those skills
An effective bootcamp will teach you the critical skills required to get a job in data science. And over the course of your study, an effective bootcamp will also force you to build at least a few unique data science projects that you can showcase on your resume. (Projects are key because they’re tangible proof that you can effectively apply the skills you’ve listed.)

Many bootcamps also offer career guidance that can improve your resume. For example, you may be assigned a mentor who can review your resume and make suggestions and improvements at the end of the program, before you enter the job market.

Will I learn the right things at a bootcamp?

Ideally, a bootcamp can take you from total beginner to employable over the span of just a few months.

In practice, it’s rarely quite that simple.

Because there’s more to data science than can be reasonably covered in a span of 12 weeks, many programs have prerequisites or require pre-learning that has to happen before the start of the bootcamp itself. Typically, these requirements aren’t too difficult to meet and the bootcamp will provide you with the resources you need to meet their requirements.

Another complicating factor is that bootcamps are typically for-profit enterprises that have to consider what sells, not just what you need to learn. That can result in curriculums that over-emphasize flashy skills like machine learning, but under-emphasize the boring-but-necessary stuff like SQL.

Missing or under-emphasized SQL is such a common-enough problem that it comes up in both student reviews and instructor assessments of some of the biggest-name bootcamps in the industry.

SQL can thus be a good yardstick for assessing whether a program will really teach you everything you need to know. When you’re considering any bootcamp, take a look at its week-by-week curriculum. If SQL isn’t included or is under-emphasized, you’ll probably need to spend some time learning it on your own to be able to get past many job interviews.

In general, a data science bootcamp should cover at least the following topics:
  • Programming in Python or R
  • SQL
  • Some probability and statistics
  • Machine learning
If you’re looking at a bootcamp’s curriculum and don’t see one or more of those elements, it’s a safe bet that you’ll have to do at least some studying on your own to achieve job-ready status.

Ideally, a bootcamp would also at least introduce you to some related workflow skills such as Git and the Linux command line — these are skills that may not be assessed on your resume or in job interviews, but that can be very helpful in day-to-day data science work.

What are the upsides of traditional bootcamps?

The biggest advantages of the traditional bootcamp approach are speed, immersion, and guidance.

Even with prerequisite work, traditional bootcamps tend to be good at getting learners qualified relatively quickly because they require total, full-time commitment. While self-study is an absolutely viable option for learning data science, very few self-learners are regularly putting in the 8 to 12 hour study days required at many bootcamps.

Bootcamp attendees also have the advantage of being immersed in the world of data science, because they’re generally living and working with other data science learners. While all-day immersion can get exhausting at times, it can further boost the speed at which students are able to progress.

Finally, in-person bootcamps typically offer personalized career mentorship and guidance. While more affordable online learning platforms often offer some forms of personalized guidance, traditional bootcamps typically offer frequent 1:1 mentorship sessions that can be helpful in preparing for and navigating the job search. Mentors can also be very useful as professional connections who can open the door to potential employers and get your resume in front of people who might not otherwise consider it.

Some bootcamps even offer tuition refunds to students who aren’t able to find jobs within a particular timeframe (although you should always read the fine print on these sorts of guarantees).

What are the downsides of traditional bootcamps?

The biggest downsides of traditional bootcamps are time and cost. Even the hybrid programs typically require a full-time or near-full-time commitment for a span of several months. Most programs also have an up-front cost in the thousands or tens of thousands of dollars.

Increasingly, bootcamps are offering ISAs (Income Sharing Agreements) as a way of deferring this up-front cost to students. But the way these options are marketed often obscures the reality: ISAs are essentially loans, and although there are exceptions, students often end up paying significantly more than the program’s up-front cash price. Here’s a specific breakdown of how ISAs cost students an extra $9,000 on average at one popular bootcamp, for example.

That doesn’t necessarily make ISAs a bad idea. And traditional bootcamps do require a smaller investment of both time and money than would be required to attain a university degree. But however your bootcamp is funded, it’s likely to require a time commitment and a financial investment that’s far higher than would be required by self-study options and online platforms.

For some students, traditional bootcamps may be totally inaccessible, since it’s not possible for many working adults to afford three full months without income (and likely at least an additional month or two on the job hunt).

Are data science bootcamps worth it?

Whether or not a traditional data science bootcamp is “worth it” depends a lot on the individual.

If the monetary and time commitments do not represent a barrier, students can benefit from the speed, structure, and additional career guidance that are offered by traditional bootcamps. And while it can be draining, many students describe the overall experience as enjoyable, since they’ve spent weeks surrounded by peers who share their interest in learning data skills.

Not all bootcamps are created equal, of course, so students should carefully assess the curriculum and read student appraisals on independent course review websites that cover data science like SwitchUp or Course Report.

Given the availability of affordable online learning platforms and free resources like tutorials, data science bootcamps are more of a luxury than a necessity for getting a job in data science. A self-motivated learner who’s found the right resources can learn the fundamentals of data science far more affordably and flexibly while ultimately ending up in the same sorts of positions (although the learning process tends to take longer).


Charlie Custer is a content marketer and marketing analyst at Dataquest with experience from everything from After Effects animation to Python programming. He spends his spare time mountain biking with his wife and pretending to be a dinosaur with his daughter.

Want to write an article for our blog? Read our requirements and guidelines to become a contributor.