Over the course of the past decade or so, we’ve witnessed the explosive rise of an entirely new paradigm in education. Sitting between the typical pathways of ‘learning it all on your own’ and ‘shelling out for a four-year degree’, coding bootcamps are ideal for students on a continuum from recent college graduates to 50-year-olds looking to make a career change.
The attraction of the bootcamp approach is that it provides the sort of crucial structure usually missing from self-learning efforts as well as an expediency and job-preparedness noticeably absent from the university system.
There are tradeoffs, to be sure. But the popularity of bootcamps is a testament to how well they can work for ambitious, hard-working attendees.
We’ve also seen a proliferation of new subjects covered by bootcamps. Once the domain of aspiring web developers, there are now bootcamps devoted to data science, UX/UI design, software engineering, and even a few focusing on building applications for the blockchain.
Galvanize has emerged as a major player, ranking alongside App Academy, Lambda School, and Thinkful. With a rigorous program, qualified instructors, and compelling statistics on job placement rates and graduate salaries, there are lots of prospective attendees vying for a spot.
If you’ve been researching bootcamps, you’ve no doubt come across Galvanize, and you might have some questions about what you can expect. I’m here to give you an insider’s perspective by relaying my experiences at the Galvanize Data Science Immersive in Denver, which I attended from September to December of 2018.
I’m going to focus on my program and my personal experiences in this article. But you should be aware that there are Galvanize campuses in other states, and that Galvanize has other programs, such as software engineering and web development.
The impression I have is that the various components are pretty well integrated. The instructors talk to each other, they mostly draw from the exact same curriculum, and so on. So while I have no firsthand knowledge of how things are done at the other locations, I also have no reason to think what I’m writing here is atypical.
An Overview of The Program
The Galvanize Data Science Immersive (‘DSI’, hereafter) is a full-time program that has you learning data science approximately eight hours a day, five days a week, for three months. If, like me, you have essentially no technical background, you’re likely to be spending quite a lot of extra time preparing and going over the material, so it winds up being more like 12 hours a day and seven days a week.
Galvanize is well-known for nurturing contacts in industry, so their curriculum is dynamic and constantly changing in response to what practicing data scientists are saying they look for in a graduate. Nevertheless, there’s still a broad structure which is consistent over time.
High-Level Structure
When my cohort passed through the DSI, it was broken up into three parts, each four weeks long:
- Advanced Analytics, which introduced us to Python and interfacing with databases, covered probability and statistics, and provided an overview of simple, predictive linear models.
- Machine learning, in which we studied the difference between supervised and unsupervised learning algorithms, and covered the theoretical basis of neural networks before building some.
- Cloud Computing, which was focused on working in the Amazon Web Services Cloud and doing distributed computations with Spark.
Each moduled finished with a capstone project that tied together and extended the learning we’d done throughout. These projects were self-designed. We would come up with a project idea and clear it with our instructors, who provided guidance and feedback throughout.
Typical Days And Weeks
More granularly, we started each Monday with an assessment that tracked and quantified our learning up to that point in the program. We worked from Tuesday to Thursday on new material, and later in the week we were assigned to groups to work on a case study. Case studies involved tackling the kind of big data science problems students are likely to encounter in their jobs, like building a framework for real-time fraud detection. On Friday, groups would take turns presenting the results of their case studies to the whole group, and would answer questions and receive feedback.
With the exception of those devoted to case study or capstones, most days were pretty similar. There were one-hour lectures in the morning — which could include visiting lecturers or advice from career services — followed by an individual programming assignment. These assignments consisted of small projects like building our own K-nearest neighbors algorithm. After lunch there was another one-hour lecture and a pair programming assignment. Pair programming assignments were similar to individual assignments, but usually a little harder and designed to be completed by two people.
The entire program ended with a kind of master capstone, a large and completely self-directed project in which we tackled a problem from start to finish. These projects were presented publicly in one of two ways: either we could give a talk or we could make and display a poster.
Personal Impressions
I can give you my own evaluation of the program on a couple of different levels.
The technical education was very good. Part of the reason I decided to take the plunge with the DSI was the endorsement of a friend of mine who works at Google. He’s one of the most technically competent people I know and he tends to have pretty high standards. Once he told me that Galvanize was ‘the real deal’, I knew it would deliver on its promises.
And it did. At ~$15,000 for three months, the DSI isn’t cheap. In fact, many people have a hard time being able to afford coding bootcamp. But the opportunity to work full-time on interesting problems is worth paying for, especially when you get to do so in a class filled with brilliant people.
Moreover, I found the curriculum well-designed, the lectures informative, and the self-directed projects were important to my development as a data scientist.
Some Challenges
There are a couple of difficulties you should be aware of. First, the DSI is obviously really, really hard, and a lot of what you learn is pretty cursory. This is a feature, not a bug. There’s no way to cram years worth of calculus, linear algebra, software development, and statistics into three months without sacrificing a certain amount of depth. I personally feel like the program’s designers did a good job of choosing what to cover and what to leave out.
There’s also going to be a really wide distribution of skill in your cohort. One of my classmates had dual degrees in mathematics and computer science, as well as five years experience doing technical internships for various companies. He flew through the material and was hired by Microsoft’s AI team in week five.
I, on the other hand, have had one college-level math class and had barely managed to do a little programming here and there in spare hours throughout my 20’s.
Again, this is a feature, not a bug. In your professional life you’re sometimes going to be the best data scientist and sometimes the worst. Being able to work productively with people at different skill levels, and being able to find some way to contribute even when you’re hilariously out of your depth, are important things to learn.
Still, be prepared to experience a little discouragement. Even when you know there’s no way you can be expected to keep up with a person who has been coding since middle school, it’s still tough to watch someone else sail through an assignment you don’t even know how to start.
Especially when you have to do it every day for a quarter of a year.
Keeping Your Head Up
If you’ve never had to practice the subtle art of remaining motivated when you feel inadequate, I recommend thinking a little about how you’re going to do that.
While you have no control over the sorts of people who will be in your cohort, if you’re lucky you’ll find them talented, hardworking, and inspiring. One of the best parts of the program was getting to learn alongside some really great people, many of whom I still see around the building and chat with.
Take the time to get to know your cohort. These are the sorts of people who go on to great achievements, and networking with them is decidedly to your advantage.
The DSI was one of the hardest things I’ve ever done, and it was made all the more so by some unique personal challenges I faced during the process. My daughter wasn’t even two yet, for example, and wasn’t adjusting well to the new house. I think I got about five hours of sleep most nights, which makes getting up and thinking hard for 12 straight hours the next day a bit more difficult.
I only relay this to let you know that even with no technical background, and even with the rigors of the program, and even with a whole lot of extra burdens most other people weren’t shouldering, I was able to finish the DSI and get a job as a data scientist.
Hopefully that makes the inevitable challenges you’ll face a little easier to deal with.
Outcomes, And What’s After A Bootcamp?
In deciding whether to take on the debt required to finance a program like the DSI you’ll want to know what your career prospects are once you’re finished.
The DSI website boasts an 83% placement rate for graduates and a starting salary of nearly $100,000. I have no doubt that this is true, but you should keep a few things in mind.
First, these data are aggregated. I don’t know for sure, but I think this means they’re taking the average of their graduates in Denver, San Francisco, and everywhere else. If this is the case that means you need to check on what the local data science salaries are like, because you’re not going to be making the same money in Denver as you would be in Silicon Valley.
Second, there’s a huge amount of variance both in starting salaries and in how quickly graduates find positions. It matters a lot what your background is like and how much education you have.
The people in my cohort who got jobs before they even graduated almost all had advanced degrees and years of technical work behind them. The same goes for those who are making on the high end of the starting salary distribution.
Since I don’t have any of these things, it took me a lot longer to find a job and I’m making a lot less money. None of this is really surprising, but it can be easy to let yourself be so seduced by the prospect of all those data science riches that you’re disappointed with how things shake out.
As before, you need to steel yourself for what the job-hunting process is like. Even when you know it’ll be tough, it’s still draining to sit around collecting rejections for six months.
All of this having been said, Galvanize graduates are doing just fine for themselves. We’re living and working all over the world, with everything from big established companies like Amazon and Nordstrom to little startups like Boulder’s Unsupervised.
Data science is one of the most promising careers in recent memory, which is why so many thousands of people are flocking to it. In a crowded space, and in the face of numerous obstacles, Galvanize has built a high-quality program that does a great job in setting students up for success in this field.
If you want to attend a data science bootcamp, I highly recommend you consider the Data Science Immersive.
About us: Career Karma is a platform designed to help job seekers find, research, and connect with job training programs to advance their careers. Learn about the CK publication.