Changing Education - How bootcamps outperform a university education

I'm @lizTheDeveloper and I work at a bootcamp. I've chosen to remain in this role long past the average bootcamp-lead-instructor burnout time - just one year. I've been doing this since 2012, the beginning of the bootcamp industry. Why haven’t I burnt out? I believe a whole helluva lot in this thing, and keeping it alive, honest, and improving is a worthwhile way to spend my life. I firmly believe that bootcamps are outperforming universities.

It's a new model for making programmers - a cross between traditional academia and the bootstrapped-learning of the tinkering children of the 70s, 80s, 90s. To me, a bootcamp is like finding another kid at school who also spends their nights nerding out, fiddling with the same toys you're fiddling with. Only now that kid isn't on an island in a sea of people disinterested or actively annoyed with your passion. Now that kid is a group of people, learning with you, fascinated by the same stuff.

The reason it's like that is this: project based learning, and relationship-based learning. Unlike traditional university programs, you have a teacher that spends all day with you, encouraging you to learn and getting to know you. They stay with you through the whole course, not like a course advisor- an actual industry expert you develop a real relationship with. They aren't lecturing to 150 freshmen, they're lecturing to less than 30 people, all of whom they've sat and pair programmed with. And what you're working on is real and it's interesting. It's something you'd find in the industry, or something it might occur to you to make for fun. You make an actual webserver with real tools that you might be using when you go get a job. You build games and make diagrams of the way the data moves between them.

Traditional Academia would never hire me- I'm just a programmer with no degree. But I've done some impressive stuff in my career- industry changing shit. The respect of my peers and my ability to do the job and do it well, along with more than a decade and a half of experience means nothing in academia. My teaching abilities mean nothing there as well- people are literally fired for doing a good job teaching at universities. The incentives don't line up- professors are incentivized to do a bad job teaching. So my career in academia never would have happened- I love teaching too much!

The way most professors start out is by wanting to do real, very scientific, unbiased research. Unfortunately, the only way for you to do that work is to go become a teacher. The skill set for teaching and the skill set for research just aren’t the same- teaching is not research. There are a large number of people who care very much about their work and are willing to do what it takes to get to do it, and it turns out that the way to their goals is through some students. On top of that, there are a very small number of roles because very little money is allocated to people doing research, and so you have a lot of very desperate, type-a, driven people trying to advance the human race- these people are the ones asked to teach you. You're in their way.

Honestly, that's why college takes 4 years. As a student, you're often a necessary evil. A lot of the work you're given to do, you're sent to struggle with it alone for hours and hours. At a bootcamp, we give our students assignments and walk them through them- first showing them how, then doing it with them, and then letting them do it alone. We also keep in touch, and coach them on their careers. They typically come back for years, always having a place to get advice, and a network of people willing to help further their career. The relationships built between bootcamp instructor and student are deeper than those between college professors and their students. The advice is solid too, because instructors come from the actual industry the students are trying to get into.

Instructors cycle in and out, for many reasons. Part of it is just instructor burnout. I blame Dunbar's Number for this. At some point, your life fills up with the relationships you have with your students. With some of them, you become friends. The admissions processes are good enough that you can authentically believe in, and bet on your students. You find yourself spending lots of time mentoring and answering emails with long and thoughtful responses. Eventually your life is full, and you just don't have time to spend on developing new relationships- so you opt the keep the ones you have and then move on from the industry. If you elect to stay, you have to compromise - relationships that form are less deep, or you lose touch with people you don't mean to lose touch with.

It's easier to form fewer relationships. This is why it's industry standard to keep instructor-student ratios low. It's good practice not to go over 1:7. But many hands for one class means that prices are high- to pay a team of developers you need some serious coin. The developers cost just as much as they do in the industry, typically around $100k, more for anyone with senior or lead in their resume somewhere, less for someone just getting started. That means every student slot needs to bring in $14,000 a year on the low end, just to break even on salary- not to mention the rent in a tech hub, and all of the other staff. This bootcamp thing is fundamentally a pricey way to teach. The margins are low, and the impact doesn't scale like hiring a developer to build an application does. 

Because instructors cycle in and out but follow the same teaching methodology and use the same materials, there's always new technology being folded into bootcamp curricula. We're teaching much more advanced concepts than when I started in 2012, but it's because the industry has advanced and obviated the need for a lot of training. Technology gets more advanced every year, but our industry believes that advanced means easier to use, and so more and more training time can be put into practice, and developing mental models for problem solving.

We bring the lean startup mentality to it, though. If went the route of DeVry or ITT Tech, or god forbid the University Of Phoenix, we would have sold 2-4 year programs of this stuff. It would cost the average student $100,000 to do a 4 year program at what we have to charge- the difference is that we think the opportunity cost is too high, students don't need that many years to break into the field. They need a growth mindset, some introductory materials, lots of  practice, and some exposure to the community. The rest takes care of itself. And it's working- most bootcamps boast higher graduation and employment rates than state universities, and a large fraction of private schools. At a fraction of the cost, several times the impact, and with the hard data to back it up.

There's also the diversity problem. More diverse students come from bootcamps, because they don't have to pay the opportunity cost associated with spending over 40 hours a week learning instead of earning. The average person can't take 4 years out of the workforce except when they are very young- if they miss that window or pick the wrong degree, they're hosed. You often don't get another go-around, and if you can't find work with your history degree or your CS education from an online university didn't actually teach you enough to be employable, then it's unlikely you'll recover. Know who tends to miss that 18-22 year old window? LGBTQ youth in intolerant households. Young moms. First-generation immigrants. Refugees. People in generational poverty. People who have experienced a catastrophe, like a natural disaster or a subprime mortgage crisis. People who are smart, wanted more, and do have the aptitude to succeed, and can prove it- but not with 4 years of their time and 100k. It takes a lot of compound interest on early life success to pay the high opportunity cost of a degree.

The signaling of a bootcamp is different than college, as well. It represents a conscious decision, rather than the pressure exerted on you throughout your high school years. Often it represents a decision you made because you're dissatisfied with how things are going, or expected to go. It means declaring "what I have now is not enough", which conveys ambition and risk-tolerance. They're the only ones capable of making it through the bootcamp gauntlet. These individuals are battle-tested, which comes in handy in the often-hectic environment of a technology company, whatever size it may be.

The future of education is agile - programs that teach you to make things, rather than expect you to pry your education from their hands. Cohorts selected for bonding and cross-nurturing. Teachers that know your name and don't forget about you when you leave. Education focused on outcomes, with programs run by educators, alongside experts, alongside students who remain excited and enthusiastic. We change fast, we base it on data, we listen to our students about what they actually need, and rapidly iterate.


25 responses
It's too soon to know, but I wonder about the longer term career consequences. For example, bootcamps don't have a general education component, so making up for high school deficiencies in the sciences, math, and humanities are off the table. This has meant less career flexibility in the workplace, at least in the past. Many (most?) students change majors at least once as they explore which degree/career fits their aptitudes and interests, learning by taking classes where they underperform or are unfulfilled. The long term benefits of quickly discarding poorly fitting career choices early in your career is valuable. We know retraining is a big factor in mid-career job changes. Will bootcamp alumni be educated/degreed enough to get into the schools that enable their next career? I also wonder about selection bias. The bootcamp student universe is tiny compared to the floods flowing through colleges. Are you getting better results because, in part, bootcamps cherrypick students? Would this effect lessen as less-perfect candidates enroll more? For all of these, it's too soon to have data. How valid are these concerns? And how might they be mitigated or addressed by the bootcamp business community?
Does your bootcamp require pre-work?
Thanks for posting this. As a tenure-track university professor that cares deeply about education, I completely understand where you are coming from. The problems that you have identified about academia are completely true. I've been surveying university introductory programming courses recently and they are, in a word, terrible. Even the successful ones offer up bizzaro or boring assignments in outdated and frustrating languages. Why don't we have first-semester students implement a ripple-carry adder in assembly language!? Or write a terminal-based game in C?! That will definitely turn them on to CS. There are a lot of reasons for this: general disdain for education at research universities, course materials that only get updated once a decade, instructors who are updating the materials whose tech stack is itself a decade out of date (peruse a few faculty web pages for laughs), and a pervasive mentality that some students just can't learn how to program. So if they don't, or don't take our courses, it must be their fault, not ours. But one reason that you don't mention is that people like you don't apply to work here. We just did a search for teaching faculty at UB. It's a decent job: low six-figures within a few years, lots of job security if you're good, plenty of useful things to build (learning tech is awful, so anything you can build yourself is better), and a few months off every summer to hack for your own account. And yet nobody like you applied. Did we require a PhD? Yes. Was that stupid. Yes. Would you (or people like you) have applied if we didn't? Probably not. So how do we attract people with your skill set and enthusiasm to work in higher education? I'll work on relaxing the PhD requirement. (It's certainly not a deal-breaker; we've employed non-PhD faculty before.) And I'll make sure that we reach out to bootcamp instructors and programs next year for candidates. Any other suggestions? And to be fair, the challenges here are huge and enticing, particularly for a technologist. We scale things. So how do you scale the bootstrap experience so that it works at a university price point? That's our challenge. It would be great if more people like you would help us out.
Sounds like bootcamps are truly progressive. Mostly because they have to be in tech because it changes every second and its impossible to produce on the bleeding edge without a constant education. Sounds like a great lab environment where people get the help they need to orient themselves to function in a practical way. The problem that I see with this is that it is we need these experiences but we need the core of all of our learning to be self driven online. And then have these experiences as we qualify in. Draw a venn diagram with all of the things that we can do in an online course to iterate through material and master it in the center. Like in Khan Academy Math. Do all of that work that has a cost nearing zero. Then data mine and match people to lab experiences, then next overlapping circle in the venn diagram. Also do all of that with no registration and payment drag. That will get people 90% of the way there and then they can jump into paid opportunities and build community connections. Imagine you just went in an answer questions and a system reported to you all of the real world opportunities and how close you were to being ready for them. People would jump in and try and close those skill gaps. We really need to remove institutional friction from the core learning experience.
21 visitors upvoted this post.