Liz Howard is a coding bootcamp veteran; before landing at Galvanize as a Curriculum Developer, she was a founding instructor at Hackbright Academy and taught at Tradecraft. And at Galvanize, she’s found a unique support system for instructors that emphasizes blended learning and iterating to keep up with changing technology. Liz sits down to talk about the length of Galvanize’s Web Development program (hint: it’s important, but doesn’t mean you’re guaranteed a Senior Developer job), how her role contributes to student success, and what Liz predicts for Galvanize’s next big curriculum change.
Can you tell us how you first got involved with coding bootcamps?
I was first involved with Girl Develop It in San Francisco before coding bootcamps really existed (in 2012). We were teaching women only, at night, for $20 per class. That was where I learned to teach programming and pairing, and I learned that you can't just lecture. Students have to do coding.
I saw students able to do so much after just a few weeks of nights and weekends classes, and started to think that if I offered enough classes, people could get developer jobs. So in 2012 I started competitive analysis to start my own school. Hackbright Academy had just started its first class, and Dev Bootcamp was on its second class. I noticed that neither of the coding bootcamps had any female instructors and after arguing with them about that on Twitter, I joined the founders of Hackbright Academy. We didn’t even really have a curriculum when we started, but we had a lot of ideas about how education was done.
How did you originally learn to code on your own without a coding bootcamp or a even a high school degree?
I am autistic, and I didn’t have a ton of friends growing up because I didn't have a great host of social skills. When I was 12, my dad taught me to code to connect with me, and I took to it pretty well.
When I was about 14, I started working on software for his startup and going to meetings with him, hearing what customers needed. I wasn't doing awesome in school and it wasn’t rewarding, but I was good at coding. So eventually I dropped out of school, when it was really cool to be a young computer prodigy.
If coding bootcamps had existed when you were learning to code, would you have attended one?
I probably would have done a summer bootcamp when I was younger and learned to code at something like a three-month Girls Who Code summer camp.
This is an angle of coding bootcamp that no one really talks about, but coding is very good for people who like to get obsessed with things. People on the Autism spectrum or who have Asperger's are able to focus on one specific thing for a long time and can get as in-depth with that subject as you want.
What about Galvanize stood out to you and made you excited to work on their team?
There's a problem in Silicon Valley, especially when disrupting an industry like education, that we don't take any of the lessons that people have already learned. I could tell that a lot of the things we were screwing up were things that the education industry had already solved. I wanted to have someone on the team from the education industry who could help us.
I went to interview at Galvanize, and I met Evan Moore, the Director of Instructional Design. Just the fact that there was a Director of Instructional Design, inspired me. Many bootcamp instructors in the industry don't have any background in education and don't have much experience, but they had an education expert on the team. I met Galvanize’s Director of Web Development, and he was very into educational theory, as well as the cool stuff we were going to teach. I realized that I had stumbled upon a little enclave of teachers as craftsmen and I got very excited about that.
You mentioned that you developed your teaching style at Girl Develop It. What is your personal teaching style and how did you find that coming from a non-traditional education?
I found that explaining the conceptual steps of programming is not always how you should teach. In a real-world job, you usually talk to your lead engineer who's running the project and they give you a task and a hint as to how to go about it. That hint could be an example somewhere else in the code base, or they tell you what to look up, or they send you a tutorial. That was where I learned the most programming – being given a task, a general guideline on how to go about it along with some code examples, and then figuring it out.
Teaching programming is like assembling machines. You show students the part of the machine they've got to install, a picture of what it looks like when it's installed, and then you set students on their way. The “figuring out” happens on their own. You can't do that part for them, and really the art of teaching is in not giving too much away at once.
I also try to show where everything fits into the overall context of programming, and how technique is valuable. After conceptual rigor, technique is really what an engineer employs.
Should a bootcamp instructor be hands-on or let students struggle?
I think being hands-on is very important. Even more, I think that you should have a relationship with your instructor who you feel comfortable struggling through something with. There are so many little parts to learning, so if you struggle through it alone, you're unlikely to hit on all those little things. Having a really intense relationship with a student is very important.
What do you think is the ideal student:teacher ratio?
In a class of more than 12, it's very hard for the lead instructor to have a relationship with every single student, which is why student:teacher ratio in these environments is so critical. It’s difficult to know what it is that students don't understand. There's only so much that automated testing can really accomplish, because an instructor has to be able to sit and watch someone reason through a problem in order to set them up with a good mindset for reasoning through problems.
That's a hands-on activity. In order to form a picture of how a student’s mind is thinking through the problem, that requires human-to-human interaction.
Does Galvanize have TAs in the classroom as well as instructors? And do you hire Galvanize graduates as TAs?
We have several levels of instructors at Galvanize. Lead Instructors run the class and have quite a bit of industry experience. Associate Instructors are either industry developers, or sometimes they are extremely good students.
There's some controversy in the industry about hiring graduates as TAs. I think it comes down to how you hire them. The biggest problem is that it helps your placement numbers, and so it’s tempting to hire students who are really nice, but not the best developers, to give them something to do for three months, then push them on their way. I’ve seen a lot of schools fall prey to that, and I think that's because the placement rate is so important to many bootcamps.
At this point, the best schools have gone through that and realized that it screws them over. It's never good to hire somebody who's going to struggle as much as your students. Once I got to Galvanize, we were past that phase. I think that everybody's learned the lesson that you can't just hire a bunch of student TAs.
What does it mean to be a Curriculum Developer at Galvanize?
There are not many curriculum developers in this industry. It’s a role that realistically gives me the freedom to do some curriculum development and some research. In order for me to do that research I have to interact with students, so I regularly come up to the classroom and check things out to see who needs help, what are they working on, and what are they struggling with. I want to know if the material is confusing or if it could be restructured.
I feel like you can't really understand how someone is learning without doing some hands-on instruction. I've never really just done curriculum development by itself, because once you get too disconnected from the students, then you’re just writing how-to articles on the internet.
Almost every coding bootcamp is bootstrapped, started with no money, and nobody spends a whole lot of time laying out the curriculum beforehand. A lot of times, schools write a curriculum as they go in response to what they know the students need to learn, and then they supplement it with outside materials and books. As a result, most schools that I've worked with are fairly disorganized, but Galvanize already had a lot of people on staff who are very good at writing curriculum. So, a lot of my job is sharing curriculum between different campuses, testing what works best, hearing anecdotes, and trying to see how we can improve the ways we present information.
It was rough, but it was very important, because at the time, especially in San Francisco, the number of Rails jobs were falling like a rock. We didn't want to take another six months to build a new curriculum and have students graduate into a field where they weren't going to use Rails. We were rewriting the curriculum as we were going, but I had done that kind of curriculum rewrite before.
Because Galvanize is six months long, how do you approach curriculum iteration?
If we're going to change actual content– for instance, right now we're thinking about switching from Angular to React– we wait until the numbers show us that hiring from React overtakes hiring from Angular.
The approach to learning at Galvanize is called blended learning, which means that students get to choose their own path. We've already built out a React curriculum, so now we're iterating on it and shoring up materials. Materials are continually being made bit better so that when students get to that point in the class, they get the latest and greatest curriculum.
What do you predict will be the next big curriculum change at Galvanize?
First, we'll probably change over to React. But the main change in the curriculum right now is to allow for more Blended Learning ie. “Choose Your Own Adventure.” In San Francisco, Node.js is hot shit, but you may be better off learning .NET in Seattle or Java in Austin.
We're trying to modularize the curriculum so that we can allow students to navigate their way through, and lean on instructors to figure out where students want to work and learn the most useful technologies. If you want to get a job at Dell and work there for 10 years, start with Java, but if you want to go get a job at a hot startup, you probably want to learn Node.js. If you want to work at a midsized company, you might want to focus on Node, or Rails, or Python. If you want to become a Data Pipeline Specialist, you can learn Java and Python and choose a Computer Science module and not focus on front end at all.
We're trying to provide enough flexibility in our curriculum, so that we can learn who you are and what you want, and give you exactly what you need and what the industry needs, rather than just having a one-size-fits-all curriculum.
Because Galvanize is six-months long, do you see your students taking more senior developer roles than graduates from the three month bootcamps you’ve worked with?
Six months does make a difference, because students get more time with mentors. Blended learning also means that students can choose their own paths, and are a little more free to go further.
Students join bootcamps with all levels of experience. I've taught students with DevOps experience, engineering experience, and Objective-C developers who want to learn web development. In every class that I've graduated, I’ve had somebody land a senior position. But that seniority is the measure of your experience as an engineer – it's not something that you can just get taught at a coding bootcamp.
What do you think successful mentorship looks like at a company that hires coding bootcampers? More importantly, how can bootcamp students choose a company that they know will be a good fit?
That's a question I get a lot from my students. Any worthwhile coding bootcamp is building a culture of learning and error and asking questions. But the reality is that there are plenty of companies that don’t believe in those values, and that may not have been the culture that people experience in their typical education. At Galvanize, we've broken our students’ brains because we teach them to expect a lot of help and a lot of guidance. Once they join the real world, they may have to reckon with a lack of guidance, and it's a very abrupt transition.
First of all is, students should know if this company has hired bootcamp graduates before and are they still around? If there are a couple of graduates on a couple of different teams, then it’s probably a good environment. Those people have been able to survive at that company, and they've probably gone through a similar experience to you. Looking for evidence of past success is probably your first step.
If there are no bootcamp graduates and you’re very junior, you're going to need a lot of mentorship, some hand-holding, and some help debugging. So you want to find a company where your lead developer (the person who's in charge of the team that you’re on) has time talk to you for at least 2-4 hours per week.
Do you think there should be a specific number of developers on a team that you choose?
That's hard to say, but you don't want to be the first coding bootcamp grad. Also, keep in mind that this is advice for the middle-of-the-road bootcamp student who is very competent and very useful right out of the gate.
If a company can give you an idea of where to start, has a fairly established code base, etc, then that's great. You’re looking for an employer who understands what you need to succeed.
I see difficulties for the top 25% of graduates, who are often placed in situations where it's very hard for them to succeed. They take jobs at tiny startups where they're the first or second person on the team, and they don't have a mentor to demonstrate how to go about anything. They don't need somebody to walk them through everything, because they can figure it out themselves. But what they do need help with is what to learn next and they need a little bit of introduction to a big subject.
How do you assess student progress at Galvanize?
The way that we assess at Galvanize is through a set of Standards of Education. A Standard is basically a task like “Build a RESTful API.” And in order for us to set a Standard, we have to have an assessment with it – a realistic task that students complete to prove that they meet the standard.
Would failing a “Standard” ever cause attrition? Can a Galvanize student “fail out?”
No, we operate based on “Mastery-based education.” So you're allowed to try again until you get it. We assume that the first time you try something, you're not going to get it.
Attrition happens usually when somebody is completely falling behind the rest of the class to a serious degree. Generally, people keep up with the pace. You can be at the back of the class, but still be keeping up, and that's fine. But if everyone else is building servers and you're not sure how to make a function work, then you either need to spend a lot more time working hard, or maybe you’re not as interested in coding as you think you are.
So we've talked about future changes to certain parts of Galvanize, but what do you see for the future of coding bootcamps as an industry in 2017?
A lot of coding bootcamps are really starting to experiment with blended learning. I think that blended learning is the future of coding bootcamps because it enables more personalized education.
Before the election, I was thinking that we would start to see a little bit more regulatory support. Coding bootcamps aren’t a replacement for college, and they’re not a trade school either. They’re a different, interesting method of education. I think that we will start to see more subjects taught through the immersive model. For example, I think of 500 Startups as a bootcamp version of an MBA. At Galvanize, one of our member companies is called GrowthX, and they do a sales bootcamp.
I think we’ll also start to see a mix of Masters programs and bootcamps. Bootcamps are socially acceptable to attend late in life, and people tend to pursue Masters programs later in life as well. If you did an 18 month MBA and a 3 month coding bootcamp, you'd graduate with a set of skills that you just couldn't even argue with. I think we'll start to compose masters programs of different kinds with coding bootcamps, and that will expand our audience segments.
It’s interesting that you bring up the regulatory environment...
I think that the regulatory environment for the next four years is going to be very difficult to predict and I think that consumers will need to be pretty savvy with what they accept from their educational institutions and how they vet them. I think Course Report will be really important to that vetting.
What sort of resources or meetups or groups do you recommend for aspiring Galvanize students in San Francisco?
We host a meetup at Galvanize called Learn To Code SF, where we do some free introductory learn to code lessons. NodeSchool is also a really great group to get started with. There are meetups that happen about once a month, and you can do it online.
Is there anything we totally missed that you want to share with Course Report readers?
One thing that I recommend is that students physically go to the school that they are interested in attending and meet the actual instructor who will be in charge of the class, because your personal chemistry is really, really important. The instructor probably has an outsized effect on your actual success in the class.