The academics I’ve spoken with about leaving academia have frequently asked how they can quickly gain the skills and qualifications necessary to get a job in technology. They’ve heard about so-called “bootcamps” that exist specifically to train people in those skills quickly. But are they worth it?
In order to get into such a camp, applicants need to demonstrate already acquired skills and knowledge. Best of all, it is possible to describe it constructively in an interview essay, to increase the chances of a positive result, you can use the help of professionals.
What you get in a bootcamp
I never attended these bootcamps, but I’ve worked with about a dozen people who did, and I’ve interviewed at least a couple of dozen job candidates whose major qualification was that they graduated from a bootcamp of some sort.
With a few exceptions, these programs are designed for people who have little to no experience in the field. They aim to take their participants from zero to professional in a very short time — usually six to twelve weeks. They are very heavily focused on practical skills, and the coursework revolves around projects that are intended to be as close as possible to real-world problems that you’d typically see in an entry-level position. Those projects are usually suitable for putting on one’s resume. Class sizes tend to be small, and there is usually a lot of personal attention. There’s also a strong focus on working in groups on these projects, which is realistic because collaboration is so important in almost any job in technology.
Participants often get very useful job hunting assistance. Some have events at the end that are like miniature job fairs, with employers coming to the bootcamp’s offices to interview graduates on-site. In fact, some bootcamp graduates are prohibited from seeking a job elsewhere until the job fair happens. Additionally, participants typically will get advice about writing their resumes and other nuts and bolts aspects of job hunting.
These are very substantial benefits to attending a bootcamp. But they come with important trade-offs. In order to get a set of projects completed in such a short time, there is a very strong focus on using the languages and tools that automate as much of the work as possible. For example, the only way you’re going to go from zero to a working web application in a few weeks is by avoiding having to learn all the nitty-gritty details of how a web application actually works “under the hood”. You’ll have a professional-looking, fully-functional web application at the end of the bootcamp (and you’ll know how to develop them on your own), but you probably won’t know any of the implementation details that have been hidden from you by the development tools you’ve learned.
To take another example, there are now some reputable “data science bootcamps”, which are flooding the market with graduates. They have the same strengths and weaknesses as the others, but the weaknesses are more severe. Data science uses some high-powered algorithms and tools that often depend on graduate-level mathematics or computer science. There is no way on earth that anyone is going to learn the foundations of those tools in such a short time. However, there are a number of high-level libraries that have done the heavy lifting for you. If you want to use some technique like “principle component analysis”, you don’t have to understand how or why it works — you can simply import a pre-existing library into your code and run it on your data. In another minute or so, you can create a beautiful visualization of the results and run some statistical tests (which you also do not need to understand).
In these data science bootcamps, you’ll get a very broad, but shallow overview of the most important techniques that are commonly used in industry. You will also get a lot of practice using those techniques on real-world data. But if all your knowledge of data science comes from the bootcamp, you will certainly not understand how they work, why they work, or the assumptions underlying those techniques. There is simply no way to gain a strong enough foundation in such a short period of time.
In the case of data science, in particular, this is very dangerous. Data scientists make recommendations to businesses that can have dramatic effects. It is very easy to lull yourself into a false sense of security because you’re using the most powerful, modern techniques for data analysis. But those high-powered techniques come with assumptions which cannot be validated automatically by any tool. In order to have any reasonable confidence in your analyses, you have to understand both the business and the foundations of the techniques you’re using. If you don’t, you can easily drive a business right off a cliff.
Who benefits?
In order to benefit from a bootcamp, you’ve got to know yourself and your limitations. As I mentioned above, I’ve worked with and interviewed many bootcamp graduates. You can split them into two categories, and it’s very easy to do so.
In the first category, there are people who are absurdly confident because they have no clue that their education has only scratched the surface of their chosen field. Their resumes are overblown lists of impressive-sounding technologies. Indeed, their experience appears to be that of someone who’s been in the field for decades. But their knowledge is shallow. If you ask for anything beyond the scope of the bootcamp, they’re shocked to discover that they don’t know the answer. They know how to do things exactly one way, in exactly one environment. If you were to take away the tools that they were handed in their bootcamp, they’d be set adrift. They don’t even know how to go about finding out how to learn.
In the realm of data science bootcamps, I’ve discovered one question that seems to completely baffle people in this first category. I name a technique that is listed prominently on their resume, and I ask them to describe a case when that technique would not work. Nothing in data science works on all problems, so this is a fair question. It’s shocking to me how few job applicants for data science can answer this question at all. But it’s crucial. If you don’t know when a given technique won’t work, then you can’t responsibly use it, and nobody should have any confidence in your analysis. Being able to use a tool doesn’t entail that you know when to use it. In my highly unscientific sample, more than 90% of the bootcamp graduates I talk to can’t even begin to answer such a basic question. And keep in mind that these are people who have listed those techniques on their own resumes!
The second type of bootcamp graduate is made up of people who are conscious of their own strengths and weaknesses. If you ask them what they liked and didn’t like about their bootcamp experience, they’ll say that they got a good start on learning a lot of practical skills, but that they’ve got a lot more to learn. They see their training as giving them a path forward as they continue their professional development. They’re curious about what they don’t yet know, and they are enthusiastic about learning. Rather than being satisfied with the skills they’ve learned, they’re more curious and intellectually engaged than they were before. These people have benefited a lot from their bootcamp training, and they can be excellent colleagues.
Beware!
Bear in mind that bootcamps are for-profit businesses, and they are not cheap. You’ll usually spend at least several thousand dollars to attend one. Don’t believe that you’ll emerge with the knowledge of someone who has years of professional experience. When you decide whether to hand over your money, be realistic, and be humble. Talk to others who have gone through the training. And if you do decide to try one, remind yourself that it’s just the first step of your career, not the last.