Screaming in the Cloud - Episode 32: Lambda School: A New Approach to “Hire Ed”

Episode Date: October 17, 2018

Are you interested in computer science? How would you like to go to school for free and learn what you need to in just a few months? Then, check out Lambda School! Today, we’re talking to Ben Nelson, co-founder and CTO of Lambda School, which is a 30-week online immersive computer science academy. Lambda School has more than 500 students and takes a share of future earnings instead of traditional debt. So, it's free until students get a job. Some of the highlights of the show include: Bootcamps were created to address engineering shortages and quickly move people into technical careers Lambda is not explicitly a bootcamp; its 30-week program gives students more instructions and more time spent on developing a portfolio Lambda also makes time to cover computer science fundamentals; teaches C, Python, Django, and relational database - not just JavaScript Employers appreciate the school’s in-depth and advanced approach, which results in repeat hires Lambda avoids the typical reputation of traditional for-profit educational institutions by being mission-driven and knowing its investors want ROI Lambda aligns its incentives with those of students; an income share agreement means the school doesn’t make money, unless students are successful Lambda’s 7-month program is less of a risk for someone later in their career; some don't have capital to support their family while going to school for 4 years Lambda incentivizes healthy financial habits; after two years of repayment, students can put that money into retirement, savings, and investments 5 Tracks Now Offered by Lambda: iOS development, UX, Full Stack Web development, data science, and Android development Mastery Based Progression System: When you're learning something sequentially, where knowledge builds, you don't move on until you’ve mastered it Lambda’s acceptance rate is around 5% and based on people who can keep up Lambda works with different partner companies to help them find qualified graduates - people they want to hire Links: Lambda School Ben Nelson on Twitter Y Combinator Wealthfront Datadog .

Hello, and welcome to Screaming in the Cloud, with your host, cloud economist Corey Quinn. This weekly show features conversations with people doing interesting work in the world of cloud, thoughtful commentary on the state of the technical world, and ridiculous titles for which Corey refuses to apologize. This is Screaming in the Cloud. This week's episode is sponsored by Datadog. Datadog is a SaaS monitoring platform for cloud infrastructure, application, and logs. Datadog does it all. Visualizations, application performance monitoring, and distributed tracing. It helps you get a grasp

on what your serverless applications are doing, as well as the infrastructure that they're running on top of. And in case you hadn't noticed, Datadog tends to integrate with everything that holds still long enough. If you're using something that doesn't have a Datadog integration, you may be the only person on the planet who cares if that thing isn't working or not. If you give Datadog a try, they'll give you a free t-shirt for your time. I've got to say I love mine. It's comfortable and my toddler points at it and yells dog every time that I wear it. It's endearing when she does it. I've been told that I need to leave the Datadog booth at reInvent when I do it. If you want to get yours, go to screaminginthecloud.com

slash Datadog. That's screaminginthecloud.com slash Datadog. Thanks to Datadog for their support of this podcast. Hello and welcome to Screaming in the Cloud. I'm Corey Quinn. I'm joined today by Ben Nelson, who's the co-founder and CTO of The Lambda School. Welcome to the show, Ben. Thank you. Thanks for having me, Corey. So The Lambda School, to my understanding, is effectively a 30-week long online, I guess you'd call it a computer science academy that has a somewhat interesting payment model compared to a lot of programs that have been tried like this in the past. So rather than me trying to stumble my way through a description, what is the Lambda School? Yeah. So essentially, the Lambda School,

it's like you said, online computer science education. And it's free until our students get a job. So we use what's called an income share agreement. And the main idea behind this is that we tried to create a more forgiving type of a student loan. Essentially, providing the ability for a student to defer payment until they're employed, and then doing it with the most forgivable terms possible. So our students, they don't pay anything until when they start. And then once they're employed, making at least $50,000 a year in a job that they got as a result of our training, then they pay back a percentage of their salary for two years. That's capped. So the maximum that they could pay is

$30,000 if they were to get a six-figure job. Gotcha. So there are a number of boot camp programs out there, but you're explicitly not a boot camp. What is it that differentiates you from, I guess, first, but causes you to make that somewhat sharp distinction? And secondly, why aren't you one? Yeah. So boot camps, they rose out of a need to huge engineering shortages that needed a faster way to move people into these technical careers. And they're pretty unregulated. It's the Wild West. And there are some good ones. There's a lot of bad actors in the space so essentially we you know we worked backwards from you know or we looked at the boot camp model and thought okay like well what's working and what's not working and you know the accelerated

model you know three months which is the length of your typical boot camp you know that that can work for some people um but but generally it only works for some of the more advanced students and um and so you know we tried to make a method that would allow more beginners to be able to have access to this type of an education. So the first thing we did that was easier is just the low-hanging fruit is we made it longer. So again, typical boot camp is 12 weeks long or 30 weeks long. So a typical boot camp is six weeks of instruction followed by 6 weeks of working on projects and building out a portfolio. So within our 30-week program, the students are getting significantly more instruction, significantly more time spent working on

portfolio development. And because we have more time, we're able to dive into computer science fundamentals. So we teach C and Python and Django and relational databases. We're not just a JavaScript bootcamp. So for our full-stack web class, we start with JavaScript. But then we go all the way through data structures and algorithms and operating systems and get much lower to the hardware. We teach computer architecture, things like that, that are designed to give the students

just like a stronger foundation, a more kind of a meaty background. Yeah, so those are some of the main differences. I would agree, having spoken to a number of bootcamp graduates myself, that there are a number of, shall we say, poor actors in this space where what they're teaching is a lot less how to do the functional job and a lot more about how to game the job interview. Where it's, these are the types of algorithm questions you're going to be asked to

solve on a whiteboard, we'll teach you to solve them, and ideally you can pass one interview out of five and that's good enough and there you go. And that's kind of a neat approach to get away from that particular model. It's not particularly common. Yeah, and the companies have been burned by that. They'll hire bootcamp grads and the bootcamp grad essentially bluffed his or her way into the job and doesn't perform well and has to get let go and then the company closes its doors to future bootcamp grad essentially, you know, bluffed his or her way into the job and doesn't perform well and has to get let go.

And then the company closes its doors to future bootcamp, you know, hires. And, you know, we've kind of seen that, you know, some of the bootcamps out there, you know, burned to the ground in front of us. And, you know, so we have to do some work to, you know, to change that perception. But yeah, so, you know, we do approach it from a more in-depth, more advanced angle, which our employers have really enjoyed. And we've measured that by repeat hires. We've had many companies who have hired our graduates have come back and hired more. And recently one was talking with us

about hiring 50 more graduates. I mean, so that's been really good to see the positive reaction from our hiring partners. So something that may wind up being a somewhat common question is when people hear school, they generally assume non-profit or they assume terrible predatory model. So you are very clearly a Y Combinator company. You're backed by a number of impressive VCs. Your Series A was led by Google Ventures. And as a general rule, not to disparage the VC model, they tend not to fund

things purely out of the goodness of their heart. They are a venture capital firm. They're not UNICEF. So how do you wind up, I guess, getting away from the, frankly, terrible reputation that traditional for-profit education institutions have deserved? Yeah. I mean, that's a whole other conversation is just how awful so many of the for-profit education institutions have deserved. Yeah, I mean, that's a whole other conversation. Just how awful so many of the for-profit educational institutions are. And they're a huge driver behind the student loan crisis. I mean, a lot of the debt, you can trace it back to a lot of community colleges and also for-profit universities.

Yeah, I mean, so our investors are looking for a return. And we are a for-profit universities. Yeah. So our investors are looking for a return and we are a for-profit institution. But we are mission-driven though. And the angle that we've approached this is, how do we align our incentives with the incentives of our students? So traditional for-profit educational school, the name of the game is get accreditation as quickly as possible so that we can get people in, siphon off as much student loan money as you can, and then graduate students and pump them out into the marketplace and just grab as much money as you can from the government or from these government-backed loan providers. And so with us, by using an income share agreement, we don't make money unless our

students are successful. And if a student is not successful in our course, so let's say they finish our course and they don't get a job, the worst thing that happened is that they spent seven months getting this free education and they're not on the hook with this debt. If they get a job and they get fired, they don't have to make payments. And so the payments are 100% contingent upon the student being employed, making more than $50,000 a year in a job that was related to the education that they received from us. And so by doing that, it means that we are not going to make money unless our students are succeeding.

And yeah, it's for profit. We'll make money off of it. But the ROI that the student is going to experience, successful student is going to experience is going to be massive. So even if a student ends up paying the maximum amount, they end up paying $30,000, that means that they're making a six-figure income

as a software engineer. And if that student still has 30, 40 years left of their career, I mean, that's a multi-million dollar swing typically in lifetime earnings that is occurring. And so, yeah, it's an investment. We do want to be a profitable company from it. But it's a very positive thing for our students because we will never profit off of failed students

and we will not profit off of a failing school. We only make money if our students are actually successful. And I think that's really powerful. And this is a little bit of a controversial opinion, but I really believe that if you properly align the incentives, then for-profit institutions can do more social good than nonprofits in many cases. And that's a little controversial, but I really believe that the way we've aligned incentives here are going to do a lot of good for our students, and it has. If we were to go bankrupt tomorrow and be dead,

we've already positively impacted the lives of hundreds of students, and that effect lasts through them and their posterity. And it's been a pretty cool experience so far. years, there was either the academic path to get here or there was the path of having worked our way up from support floors, working in data centers, getting racks falling over and crushing us. And we learn over time that different skills that help us to evolve into the modern quote unquote cloud world. And the question now becomes with those jobs having been gone, what are the viable paths to get to this type of career? So if I understand what you're saying, you wind up charging a percentage of the student's income after they graduate. You take nothing until they make at least $50,000 a year, and you hard cap

at never taking more than $30,000 versus, I guess, more traditional approach of a four-year degree and, I don't know, underwater basket weaving, where suddenly it's going to cost you $120,000 to get it, no guarantees tied to it at all, and that debt will follow you until you die. That's a very different model. Yeah, yeah. I mean, it's riskier. When you say it's riskier, you're talking about Lambda School or you're talking about the traditional path? I'm talking about the traditional path for somebody who's already later in their career. And so, yeah, the idea here is that, you know, with Lambda School, we try to, you know, de-risk that transition as much as possible. So back in the beginning of your question, you talked about where is this next generation coming from, this next generation of technical talent

and software engineers. And traditionally, it's had to come through universities and these four-year programs. I've checked. You cannot major in serverless. You cannot major in React. You cannot major in writing web applications. It's weird because people start talking about vocational abilities and going back to academia,

which is explicitly non-vocational in nature in most respects, and hoping for that to have the answers. And it's never felt to me like that's a viable path forward. Yeah. I mean, you talk to the university and you'll ask them, what's your purpose? And the purpose is that usually you get some answer about how it's creating a more well-rounded person. And it gets into more of these intangibles. But typically, from the student perspective, they're there because they want a good job. They want a good career. And it's the next step forward. And so there's a little bit of

an incentive alignment. And again, we're not anti-university. And that's great. But it just isn't an option for a lot of people. And so anyway, that's where we step in. One of the hiring partners we've been working with, this is a bigger company, top of the Fortune 500 list. And they project that within a few years, that they alone will have enough job openings to hire every single computer science graduate in the entire country from every university. And I mean, it's a pretty bold claim,

but that's the type of scale that they're operating at. And there just is not enough talent in the workspace. And so part of what we're trying to do is fill that vacuum. And with it being a 7-month program, with it being online and accessible from everywhere, and because you don't have to pay upfront, you only have to pay if it's successful for you, that de-risks it for a lot of people that are later in their careers. Maybe they're married and have a family now and they live in the Midwest or something, or they're not near any big technology hubs or don't have access to a good university or don't have access to... They just don't have the capital to support their family while they're going to

school for four years. And so the idea here is to create this vehicle to where we can take people that are in industries that are maybe being replaced by automation or just people that are in a job with stagnant wage growth and they want to get into something that has a higher earning potential. And by making it online, by having an income share agreement,

by making it seven months long and intense as opposed to four years, it facilitates a quicker movement. There's also something to be said for the repayment starts as soon as they land a job that pays above a certain percentage, which in most cases is more than they've made previously. To some extent, you're enforcing good financial habits of, at that point, they haven't had the chance for lifestyle inflation to really set in to a point where if you pay someone $100,000 a year and there's no strings tied to that, people start to spend $100,000 a year in many cases. If you're entering into this with a percentage-based approach, to some extent, you're almost incentivizing good financial habits because after their two years of repayment are done, they're used to living without that. They can start putting that money into retirement, into savings, into investments, and I guess coming out with a better financial health model from this. Yeah, exactly. I mean, and that was a cool insight that we had. And so we've actually

partnered with Wealthfront. And so, yeah, I mean, and that's what we tell our students is, when your payments stop, just pretend like you don't have that increase in money and just take that same money, transition that, start investing it. And so we have some people that work with our students and talk about safe, conservative ways to invest and save for retirement and things like that. And yeah, you're right. I mean, it leads to a good financial lifestyle that can benefit the students throughout the rest of their life. What I find interesting about a lot of these stories is you talk to people, in many cases, who've been out of college for 10, 15, 20 years, and they're still being haunted, in many cases, by student debt. I've looked at your offering. I have a hobby of playing around with personal finance more than I probably should.

But it seems to me, no matter how I look at this, there really isn't a part of the story where someone winds up taking your course and winding up down a financial rabbit hole. The only question comes down to the opportunity cost of, is this the best program for an individual student? But it doesn't strike me as there being any edge case or corner case where doing this could be financially disastrous for someone. Am I missing something obvious? No, yeah. I mean, that was the intent from the very beginning. The only thing that our students need to risk to come to Lambda School is the opportunity cost.

The seven months they're spending in the program, the months they're spending on the job search, and the time they're spending to prep for the class. Which that does add up to a lot of time. And we're working to make sure that our hiring rate is as high as possible. But yeah, we never want to take money from a student who isn't able to afford it. We don't want to take money from students that did not greatly increase their lifetime earning potential. We don't want to take that money unless it's fair,

unless it's benefited the student. Talk to me a little bit about your curriculum. You have multiple paths or tracks people can go down. What are those? Yeah. So we teach iOS development, UX, full-stack web development, data science, and then Android development. So right now we're starting with these different technical tracks. Really, we see ourselves in the business of taking someone from income A, increasing it up to income B,

and then we take a piece of the delta. So really, eventually we'll expand beyond that into other non-technical areas. I mean, so we've even considered potentially a nursing school, things like that. But really, it's moving people into jobs that have high demand behind them. Yeah, so those are the five tracks we have right now.

And then that will expand as we go forward. One thing that I find neat is you have this, I guess you described it as a mastery-based progression system. What does that mean? Yeah, so this is something that's widely acknowledged as probably the best way to learn things and and it's something that most schools can't uh they can't do it at scale effectively and so essentially mass mastery based progression is the idea that you know when you're learning something

you know sequentially where you know where the knowledge builds on you know previous topics you don't move on to the next thing until you've mastered the thing you're on. So the idea is like, you don't learn the next topic until you've mastered what you're currently studying. And so what this means in our course is that when we're teaching... I'll talk about our web course. That's our biggest one enrollment-wise. So I'll use that specifically here as an example. But with JavaScript, if a student doesn't master our JavaScript lessons, then they don't move on to the React lessons. So like we're not going to try to teach them React if they don't know JavaScript.

You know, we're not going to get into, you know, some of the more advanced, you know, database, you know, design lessons, you know, if they haven't, you know, mastered, you know, some of the basic API principles. And so the idea is that every week, a student takes an assessment. And if they don't pass with an 80% grade or more, then they repeat the week. And we do our best to remove any type of stigma around failing. And the idea is that the students will continue to spend time working on

the topics that they struggle with before they move on to something else. And we found that students that before would have become overwhelmed, would have wanted to maybe drop out. Instead, it gives them a way to succeed and it really empowers the students and improves the experience significantly. One question that I have for you is, I don't know if you're public about this or if you can give me numbers, but what does your acceptance rate look like? Something I noticed on your site is it's not sign up here. There's an application button, which implies at least some subset of people are not a fit for your program.

Yeah. I mean, unfortunately, we have to filter because it's a seven-month program. It's accelerated. I mean, hour for hour, our seven-month class comes out to being a little bit more than about a year of college is how you can think about it. Accounting for homework time and class. But it's taking a year of college and condensing it down into 30 weeks. And so we move at a quick pace. So we need people that can keep up. Yeah, so our acceptance rate, it fluctuates, but it's sitting around 5% right now. But our attitude towards it is we don't want to be exclusive.

We don't want the exclusive nature to Lambda School to be part of the value that we provide to students. We want to accept everybody who can meet the high standards that we set. So think of it as like if Stanford suddenly opened their doors and everybody who applied who had a 32 on the ACT and a 3.8 GPA could get in. It's a high standard, but anyone could attend if they could meet those standards. And that breaks down with their type of a model. But it is a little crazy that they turn away 95% of the people that want to pay them hundreds of thousands of dollars. But so our perspective is, we want to provide this opportunity to as many people as possible, provided they can keep up and they can do well in the course.

Because again, we don't want them to push through the course if they're not going to succeed and if they can't keep up. Yeah, so we have... It's like a tryout at the beginning. So the students apply, and then they have to pass an entrance exam to be able to start the course. And then we have all these free classes that we're running with live instruction that are designed to get people up to speed so they can pass the entrance exam.

So the entrance exam for the full stack web class, it includes things like our students need to understand what variables are. And if statements, for loops, functions, all in JavaScript. And so it's nothing too complex or too terribly difficult but it's you know it's about what you would learn in in the first semester of it like an intro to computer programming class in college and so there's the idea is that we don't start from zero um on day one of classes as all the students they know a little bit we've proven that they can you know that they do have an aptitude for the material and that they can be successful

with it um and then we then we push them through the you know the that they do have an aptitude for the material and that they can be successful with it. And then we push them through the, you know, the intense seven-month course after that. So let's say hypothetically that I find what you've said compelling. And as, let's pretend for a second that I work for a giant company. Spoiler, I'm unemployable. But how would I go about hiring your graduates? I mean, Is it something I can do other than keyword resume searches? Do you have a reach out here and we'll put you in touch with our alumni style program? Or is it still early days to the point where that hasn't been formalized as yet? Yeah. So yeah, we have a hiring partners page on our website. And we have a guy who his full-time job is working with different partner companies to help them find qualified graduates for their company and what they're looking for.

So because we're online, we have students. Right now, we have close to 600 students currently enrolled, and they are scattered all over the United States. I mean, every metro area, we have students everywhere. And so then what we do is we look at students that are within that geographic region. And we look at the requirements for the job. And we work pretty hands-on with these companies to help them find people that they want to hire. And this is a free service. It's a free white glove service.

I mean, we don't want to or we you know we don't want to put up any barriers to people hiding hiring our grads you know we'll you know we do uh we do quite a bit to work with companies to make it easy for them to find qualified students to hire and um you know and make sure they have a good experience we follow up with the companies and we get their feedback on the student and we apply that to the curriculum and and uh you know we really work with them closely to um you know to gather any type of feedback they have around the students specifically and how that could maybe reflect on our school. And if there's anything we can do better, because we're constantly trying to improve the curriculum, improve the way we do things, and we're always in this high learning mode. One thing I feel the need to point out is, despite how this entire episode has played out, this is not paid placement.

You're not sponsoring this show. This is something I'm legitimately interested in. And it sounds to some extent almost like I'm trying to promote the school itself. That's not based on any financial compensation. I just really like the idea personally, which is why I thought this would be a terrific episode. Thank you. Thank you. I appreciate that. I just really like the idea personally, which is why I thought this would be a terrific episode. Thank you. Thank you. I appreciate that. And, you know, really happy to be on here and talk about Lambda School. And yeah, always fun to find another fan. Absolutely. People can find out more at lambdascool.com.

Yes. Yep. That's the best place to go. Terrific. This has been Ben Nelson, co-founder and CTO of Lambda School. I'm Corey Quinn, and this is Screaming in the Cloud. This has been this week's episode of Screaming in the Cloud. You can also find more Corey at screaminginthecloud.com or wherever fine snark is sold.