Bootcamps don’t make High-Caliber Data Scientists

The (not so obvious) Drawbacks and Shortcomings

Bootcamps are a legitimate option for education. Many have produced Data Science graduates that were able to secure various job opportunities. However, there are significant drawbacks to consider

Bootcamps are geared towards turning students into adequate candidates for entry level roles, usually at startups or small to medium sized companies . They are not geared towards making you a candidate that can easily land a role at a FAANG company( F acebook, A pple, A mazon, N etflix, G oogle) or a unicorn startup like Uber or Airbnb.

Bootcamps will advertise in a way that suggests many of their graduates place into top tier tech companies, however, this is very disingenuous. Graduates that do land in those top tier roles should be considered as outliers and are not to be expected as the norm .

Generally speaking, bootcamp grads that are able to land in a FAANG role are able to do so because they have much more on their resume than a simple certificate of completion. They may have additional work experience than the average bootcamp student(the average student is only 28 years old) and additional formal education. Additional experience in data management, consulting, writing and speaking are also highly sought after skills for Data Scientists.

Always remember that bootcamps are a business enterprise. Though bootcamps can technically claim their curriculum produces FAANG caliber Data Scientists, the reality is that their curriculum is not the only the reason for their students’ success.

Bootcamps are notorious for cramming large amounts of information into a short period of time and students are expected to not only keep up, but also master the material . Generally speaking, most camps are somewhere between 3 to 6 months in duration and demand a full-time commitment(40 hours/week, if not more).

To be considered hireable for an entry level role, a candidate must have the following technical skills:

• Proficiency in Python or R (Python is more sought after)
• Proficiency in SQL
• Proficiency in statistics, linear algebra and calculus is a nice bonus. Additionally, candidates should be familiar with creating, running, and evaluating A/B tests utilizing various evaluation metrics.
• Proficiency in Data Visualization with Matplotlib, Seaborn, Pyplot, etc. (Tableau is a nice bonus but not necessary)
• Proficiency in both Supervised and Unsupervised Machine Learning techniques. Experience with pipeline development, model evaluation, and optimization.

This is the majority of the curriculum a bootcamp will provide which is a ton of information to cram into a few months. Assuming you can keep pace and achieve a level of proficiency in the above skills, you’ll just barely qualify to be an entry level Data Scientist. However, there is sooo much more to learn.

These are the additional skills most employers are looking for (depending on the industry of course :))

• Natural Language Processing
• Deep Learning
• Time Series Analysis
• Big Data Analytics
• Cloud computing
• Biostatisics
• Robotics and computer vision
• And more!

Again, this is an immense amount of additional material to learn. There is no way a bootcamp can provide an all-encompassing Data Science education in a shortened time frame. It simply is not possible, there is way too much to learn.

In theory, you could learn the additional skills after graduation. You could also spend more time relearning and mastering the core curriculum to become an even better scientist.

But… if you’re going to need to do months of additional self-teaching, wouldn’t you have been better off self-teaching yourself the entire time?

• Monetary and Opportunity Cost

The majority of people interested in pursuing Data Science will not have the resources or the time to commit to a bootcamp.

Bootcamps are very expensive. Tuition itself is thousands of dollars and I have seen camps charge as much as $15,000. In addition, bootcamps do require a full-time commitment of at least 40 hours, with additional time outside of class spent attending study hours, completing homework assignments, and attending networking events.

There are some camps offering part-time enrollment, where the time commitment is 20-hours per week. But again, there are still time-consuming activities you need to be cognizant of before enrolling, especially if you are planning to continue working full-time during enrollment. When you do the math on that, if you do enroll in a part-time program while also maintaining a full-time job, that’s easily a 60–70 hour work week for your entire duration in the bootcamp.

• The completion certificate is a weak signal to employers

Picture this: After all the months of working tirelessly through your bootcamp, you are finally awarded a certificate of completion. You begin your job search only to find out that employers are not lining up to set up interviews with you. Why not?

The bootcamp certificates of completion only signals that you completed coursework from a bootcamp. It does not signal you will be a competent scientist.

A certificate of completion will not be the reason you are given an interview or an offer over another potential candidate. At the end of the day, Data Science bootcamps are incredibly new and have not been around long enough for their accreditation to be meaningful enough for a hiring manager to choose you for an interview or role.

What’s going to set you apart from the crowd are the projects on your resume that exemplify your proficiency in the technologies an employer is looking for. A strong portfolio of projects will get you an interview and performing well on the subsequent technical assessments and in person interviews will land you the coveted role of a Data Scientist. There is nothing a camp can provide for you that you cannot provide for yourself.