Bootcamps should be more transparent | Hacker News Reader

Bootcamps should be more transparent

medium.com

105 pointsDangeranger64 comments

As someone trying to hire web developers, it looks to me like Bootcamps are more like finishing school than trade school. Candidates that come out of a bootcamp have learned how to signal membership in the Silicon Valley in-group.

Their resumes have all the right keywords, and refer to "projects" they've worked on, built with all the frameworks and tools currently in fashion. They've practiced their interview technique and can talk confidently about continuous integration or opine about React vs Angular or describe how well they cooperate with their teammates.

If they don't get the job, they'll politely ask for advice on how they can do better, and maybe ask for a coffee date, so they can learn about the industry and how to find a mentor.

If you don't look closely at the resume or ask explicitly in the interview, you might think they have actual work experience, but they don't, it's all just coursework.

So I'm a bit ambivalent. Yeah, we desperately need more developers in the tech world. Yeah, startup culture is ridiculously tribal and unfairly biased against those who don't know the secret handshakes. But 3 months does not a developer make. When I'm hiring, I now look very skeptically at candidates that have been to a bootcamp. Experience has taught me that interviewing them is usually a waste of time.

> So I'm a bit ambivalent. Yeah, we desperately need more developers in the tech world. Yeah, startup culture is ridiculously tribal and unfairly biased against those who don't know the secret handshakes. But 3 months does not a developer make. When I'm hiring, I now look very skeptically at candidates that have been to a bootcamp. Experience has taught me that interviewing them is usually a waste of time.

That's all the more reason these bootcamps should be way more transparent than they are.

There are people for whom the cost of attending such a bootcamp is enormous - they might spend their entire life savings or finance the attendance of such a camp, putting them in a bad debt spiral. And while many of the camps are selective and only hire people with previous development experience, I'm sure there are some with more 'flexible' standards so to speak that will happily take your money, regardless of your background.

One of the things they could do to make it more transparent is to signal that software development may not be for everyone - just like being a doctor or an athlete or some other profession that requires extreme passion and dedication.

> Experience has taught me that interviewing them is usually a waste of time.

So we have a situation where hopeful individuals spend a ton of money on questionable bootcamps for the promise of a shot at a job... and the employers think interviewing said individuals is a waste of time. That's a lovely thought. :/

It's the same thing that happens with tech certifications which are providing a similar service. It often doesn't tell you anything unless you happen to know and are familiar with the ones that have value. It does make it harder to read a resume or get a good feel for a candidate. They may be able to say the right things but without the true insight that makes it valuable. It often can land you with someone who understands the motions but didn't develop the intuition. It's one of those weird self selecting groups- the people who think they need a bootcamp or a technical certification to get the job might not not be the right type of person to begin with, is instructed the wrong sort of way, or ends up with the illusion that they are near the end of their learning rather then just starting at the beginning.

On the other hand there are people who really make the most of it and come out extraordinarily better. But those top of class folks are by definition a minority. Other people succeed, but don't exceed. Other people pass. It's the same reason why some people devalue college graduates.

What's wrong with that thought? That seems to be the interviewers genuine experience.

IMO, programming is learned through repeated banging your head against hard problems for a long period of time. 3 months cannot achieve this. My exp has shown me that bootcampers are well-versed in hype and have built applications under perfect conditions, which as a practice, has little value.

Yeah, we desperately need more developers in the tech world

Well, management needs more cheap workers early in their careers who can be easily exploited with the prospect of 0.01% share options, but that's not the same thing. If there genuinely was a shortage of workers, there would be no ageism, but it's rampant. Zuck stands up and says "young people are smarter" and no-one bats an eyelid...

See the sentence right after that one. In the best cases the Bay Area tech community is cliquey, in the worst case it's illegally discriminatory. The recent fashion for founder-friendly term sheets hasn't really extended to employee-friendly employment agreements. Sure, I get that.

And yet, it's also true that there aren't enough developers. I've dealt with hundreds of candidates over the past year. There are a few who rejected an offer, presumably because they had a better one some where else. Most, though, are not qualified, and most of those that are, accept an offer somewhere else before completing their interview.

As an experienced engineer, I'm happy that it's a seller's market for tech labour. Someday I'll be on the other side of the table. But damn, there are so many things we could accomplish if there were more of us.

I was intrigued by your comment:

"The recent fashion for founder-friendly term sheets hasn't really extended to employee-friendly employment agreements."

What do mean by "founder-friendly term sheets" and "employee-friendly employment agreements"? Besides the "what" of those two thing how do those things relate? Or how should they?

Thanks.

Love this. Some bootcamps are doing the world a great service, but there are also an increasing number of sharks in the water.

I run a 4-week "pre-bootcamp" program to help advise students on making the right choice. I hold my classes in the evenings so students can try out coding before they make the HUGE commitment of quitting their jobs and shelling out 10-20k for a bootcamp.

Sometimes, the right choice is to not try to be a coder. People get excited about six figure job prospects, but the truth is that coding isn't for everyone. The right approach is allow students to "fail fast," just like a startup. I do believe everyone should at least be exposed to coding so they can make an educated decision about it though.

On a slight tangent, I'd love to see more coding schools putting their money where there mouths are by offering income-based repayment (IBR). I've worked as an instructor at a coding school offering IBR, and I've seen first hand how much of good thing it is for everyone when success incentives are aligned.

If someone would have told me I would suck at coding two years ago, I'd be thanking them immensely.

Interesting. Would you mind elaborating?

I met a guy who was looking into a bootcamp. He was over forty.

I immediately thought, I don't care if he's the best student in the class; he probally will have a horrid time getting a job.

I don't think I've ever seen an industry that's so guilty--actually I won't say the word, but if I was offering a bootcamp, I would just mention the game is geared for youth.

I'm not saying you can't take the courses.

Actually, does anyone know anyone who has taken one one of these bootcamps, who's over forty, who got, and kept a job in tech?

(The older guys I know, interested in programming/web developement all had to work for themselfs. I know one guy who made a killing. The rest are barely supporting themselfs. None went to a bootcamp. They couldn't justify the cost. I don't know if they did the right thing, by not going to a bootcamp.)

I like the idea of transparency from these bootcamps, it'd be a great way to vet them before committing $20k and several months of your time. It makes sense that bootcamps wouldn't be willing to divulge that information though, especially if their numbers aren't great.

But it feels a little unfair to hold the bootcamps completely responsible for people not being able to find jobs. Seems like people have this idea that just showing up is enough. It's not enough to just show up with a piece of paper like 'hey I finished a bootcamp/degree, gimme a job!'

I got my first job out of school with a combination of luck (knowing how to implement conway's game of life) and passion (a side project I had just shipped). Most every job after that has been because of passion. Whether or not you went to school for CS (or at all), or finished a bootcamp is completely irrelevant imho. I probably know more about von neumann architecture, big O and sorting algorithms than my friends without CS degrees... but I've worked with or admired plenty of people who never took a CS class and have mastered lots of things that I haven't. And I can definitely recall a few friends in undergrad who I was sure would go far and never found a single job cuz they sat around waiting for Larry Page to call them personally.

So yeah, tl:dr; school doesn't matter all that much if you're not motivated to continue learning. And if you don't enjoy programming enough to learn && experiment/play you're probably not gonna do well.

I agree that its not fair to blame the bootcamps entirely, though I still think releasing better statistics might allow people to make more informed decisions, or at the very least weed out some of the shadier bootcamps.

That said, I think your tl;dr sums it up pretty accurately. If you're not that into learning something, its going to be hard to make a career out of it.

I almost went through one of the bigger bootcamps in Austin a couple of years ago, and I'm so glad I didn't for several reasons.

Everything about the application process felt like a for-profit university: they constantly blasted their impressive placement and salary figures at me, making it sound almost like I was guaranteed a $100k job right out of the program. A 96% placement rate and an average starting salary of 100+k allows them to paint a certain picture in applicants heads, and most people have no idea how hard it is when they sign up.

Funny enough, this bootcamp had all of their current students use the same job title on LinkedIn to show the bootcamp as work experience, which made finding alums VERY easy. I sent several alums who didn't get jobs quick messages asking about their experiences, and the responses were eye-opening. They were coached on how to go to networking events, where they wound competing against all of their classmates for junior gigs. After a few months, they went down in the bootcamp register as "failures", so they didn't count towards the glowing placement figures.

But the icing on the cake came years later, when this same bootcamp contacted a friend about recruiting some of their students. During the conversation, they disclosed that they were a contingency recruiter for their students: these junior candidates were going to cost the company six figures with the recruiting fee, which was crazy.

These students shell out $20+k for these programs and become unattractive candidates because the school wants a recruiting fee for someone that's been coding for 12 weeks, and the students probably don't even know what's going on.

There's absolutely a need for transparency here.

Name three great for-profit colleges.

Ok, name one.

Now you see the average for-profit bootcamps being acquired by traditional for-profit colleges.

The duty to make financial returns for investors necessitates extracting money from students. Like an ecosystem with the oxygen slowly extracted out, the good things happening there eventually die.

I believe the only two non-profits in this space are Ada Academy in Seattle and our Turing School in Denver. At Tuting, we believe in a world where the people building technology represent the people using it. We'll still be here in 10, 25, and 100 years working towards that vision.

Code bootcamps have become the modern ITT and Phoenix Universities. Seriously, you DO NOT need to go to a bootcamp. When I hire and someone tells me they're from some "prestigious" bootcamp, do they expect me to believe they've become awesome by going through one?

What did you do until bootcamp? Show me you had initiative on personal projects. Show me non-trivial projects on Github. Show me pull requests that you can read and understand other people's code. Show me submitted issues that you diagnosed a bug and gave enough information for another developer to reproduce it. Github tells me a lot more about you than graduating from a bootcamp.

I'm a graduate of an after-hours web development bootcamp who wants to move to a full-time web dev job.

I certainly can't move to a full-time web dev job at a reasonable salary after the bootcamp I attended. If I was willing to intern for 6-12 months at minimum wage, I would be able to - but I could have done that without the bootcamp.

The bootcamp doesn't share stats, only a post-graduation "employment" percentage. Which is pretty meaningless, for example, I'm employed but not doing web dev.

Every after-hours bootcamp I've heard of has had awful reviews. You're missing out on the whole "immersive" part of it, which is really what matters in a lot of ways.

Like how you learned Spanish a couple days a week in middle school. Do you remember more than how to count to ten and ask where the bathroom is? No, but if you had to live in Mexico for 3 months you'd get pretty adept and be able to make your way around and likely have conversations, assuming you put in the time to actually learn.

I would have preferred to do an immersive, in large part for these reasons, but couldn't afford both the tuition and loss of income for months.

The BC I attended claimed to have a flipped classroom model, which would have been better that in instructor in the front of the class demonstrating code. In addition, I was apparently one of the few students who actually did want to change careers, so my "peers" were never willing to work outside of class. Despite my day job, I put in at least 2 hours outside of class for each how in class.

Even with the after-hours approach, if the classroom model had actually been flipped, and there were several peers to work with outside of class, it would have been significantly better, and I think I could get a junior dev job.

I think you could still get a junior dev job, if you just do a bunch of projects, even if trivial or done before, that demonstrates you know what you're doing. If you don't know what you're doing, the projects will help you learn. Win-win.

If you don't mind me asking, what program did you go to?

Your main competitor that was just sold.

Your feedback is exactly the same that I get from others coming from that program. Thanks!

I like how forthcoming you are in your report, and agree that all bootcamps should do the same.

Perhaps we'll see an effect where bootcamps that do high quality reporting will actually be chosen more frequently and thus push other bootcamps to do the same.

Title is a bit click baity, but I agree with the author's points.

He sees it as a negative thing, but I see it as a massive improvement on college/uni/grad schools, who wouldn't even know where to start with gathering/publishing this data.

Bootcamps are subject to market forces in a way most colleges are not, and it's initiatives like this (what the author is proposing) that give me a sense of optimism for how they'll evolve.

> He sees it as a negative thing, but I see it as a massive improvement on college/uni/grad schools, who wouldn't even know where to start with gathering/publishing this data.

This isn't true in the US. Many schools do actually try to collect this data, but they're typically dealing with far larger populations with far less response rate to surveys, so it isn't as readily available.

Law schools especially report the number of students with jobs, but they're notorious for gaming the system by doing things like hiring their own graduates for a year post graduation and similar nonsense.

True - many schools collect employment data, but unlike the author, they don't tend to shine light on "total costs of program" (like four years of $30-50k/yr salary + housing expenses + books)

Ah, law schools...

This is a great point. Can you imagine if every college major published its job placement stats and average salary?

Thanks for actually getting into the details on graduation outcomes. I've considered doing a bootcamp but I've been held back by a lot of uncertainty about graduation outcomes. Being able to go through the raw data and find people like me is really helpful.

I would be interested in knowing more about the background of people before they joined. Do they have a STEM background, what was their job title before, how old are they/how long have they been working, etc. Also if you could categorize the types of companies they work for (startup, large tech companies, non-tech companies, etc.) that'd also be interesting to see.

Thanks for providing transparency to your program!

Something else to consider - how sure are you that you enjoy coding and it's something you can be good at? I highly recommend taking the time to evaluate this before you commit to a bootcamp as well. Feel free to email me (address in my profile) if you'd like me to elaborate.

I went to a bootcamp, I have a pretty varied but useful skillset in growth hacking, front & back end dev, the tech industry & tech trends, software & frameworks ect. I'm unemployed, however I attribute this moreso to moving from LA to a tiny east coast town, ect. I am just being transparent about my bg to give context. Also, I have had freelance & worked at a startup, ect. I finished General assembly ~2 years ago.

I agree with the author but the issue is that it is both very difficult to measure, track and maintain data and it varies wildly. Consider it almost like a vc portfolio. some years, everyone applying to bootcamps will get a job because of timing & numbers ect. However, more bootcamps spring up, and pre existing ones do more cohorts. All of a sudden, there aren't any "hot companies" to invest in(to carry on the vc analogy). You have to keep putting that fund to work though so the pool becomes diluted.

Another issue is accounting for massive disparity within schools & within the wider ecosystem. Some people in my class were very talented & had a lot of peripheral experience. Some codecamps are not selective. I taught myself html/css/js a bit of rails and jquery ect BEFORE, i ecen applied. I laugh at even typing that now as it is trivial, but some students had never even looked at web inspector. not only does that change demographics, but 1-2 students can kill productivity here.

This is why unoversities evolved into what they are. Some schools carve out a niche as elite and their rep allows them to be selective. sone don't, because 10s of $1000s per applicant is crazy to just add 1.

everyone knows this so i wont beleagure the obvious anymore, however, the last bit is tracking. I have a job interview tomorrow, and have been working on my own startup ect. I am financially unsuccessful and sonewhat professionally so. However, I consider it a success in that I can hacn out projects and build shit and as long as I can eat, I love it. Other people got jobs for 2 months right out of bootcamp, were terrible and got fired. Schools with "great" numbers have no idea how they actually look outside of that snapshot. its tech. and its a first job. people have freelance gigs and then go to 1-3 conpanies within a year or so.

tl/dr

transparency is inportant and I am for it 100%. However, if you cant/wont be, then be selective and protect your brand and rep. Being a long term institution that is selective & well known will help you retain teaching talent, attract top applicants and be worth way more than the short term $$$.

> the issue is that it is both very difficult to measure, track and maintain data and it varies wildly.

Also went to a bootcamp and I agree with this. I think the variance in ability between students is greater than the variance between bootcamps. People come in with different levels of experience and backgrounds. People are looking for work in different places.

I think reviews from former students are a better solution than just numbers.

Reviews and speaking with someone who attended is huge, and I agree. My cousin went to a different campus than I did General Assembly NYC and had a much different experience. Ryan Dahl, used to hang around at NYC and even did a talk on using Node.

It really is exactly what you said. Speak with teachers, former & current students and read reviews.

How was your exp. btw?

Actual data: http://report.turing.io/

Speaking as a hiring manager, I place a much higher value on verifiable project experience (either I can see it or you can walk me through your unique/non-trivial app in our interview) than any credentials. I've seen a lot of candidates who are focused on credentials crumble in the technical interview.

Granted, there's a huge universe of people out there who are more than willing to take your money (including Microsoft), but the really good people I've hired blew me away with their project descriptions and not their list of certifications.

/s/guys/people

I'm a cofounder at Hack Reactor, and I also attended this meeting. (Sup Gregorio!) Here's my 2c on the 3 things he says we could not agree on:

1. "All programs must report their graduation rate and what constitutes a graduate (not only their completion rate)." I can't remember this being a topic, so I feel like I must have missed something.

2. "All programs must report how many graduates are employed as full-time software developers, and separate other tech-related jobs, and non-full-time forms of employment from this metric." Gregorio is right about this: bootcamps should be transparent about this. Reasonable people can disagree about whether a bootcamp should count eg "growth hackers" in employment figures (Hack Reactor would not, per our methodology[1]) BUT bootcamps should DEFINITELY make that data available. This was shot down.

3. "All programs must make available the raw data that they used to base their metrics." Again, I can't remember when we talked about this. It's kind of a good idea, but unless you have robust non-employment data, an anonymized data dump is a csv with like 3 columns. Ours just had salaries, outcome codes, and time-to-hire (which we already published in histogram form), so we didn't bother publishing it. (Turing's has lots of fun demographic data that would be out-of-scope for an industry standard about employment.)

Do you find that every student wants to be a full time developer though? What if the student were fully capable of finding a job as a developer but that's not what they wanted to do? Or maybe they were just open to many possibilities. Seems like just have a simple statistic like % of students that became devs would be 1. misleading and 2. acting against the bootcamp's own interests.

In my mind including "tech-adjacent jobs" is a more interesting number. These programs attract a lot of professionals who are trying to add programing to their resume, but not necessarily become full time code monkeys.

A designer who can write html/css/js is more valuable than one who does not. A PM who cannot code is a nightmare to work with. Working with a marketer who can hook up their own jquery plugin is a godsend.

Why not include both sets of stats? There are a large number of bootcamps these days and I obviously cannot speak to all. However, for the program I attended, I'd say 90%+ of students were interested in full-time dev jobs. I'd want to know if most post-course jobs were not highly technical.

Aside from that reasoning, I see this as a means to differentiate different types of programs, if they do in fact exist. I don't actually buy that argument; while I obviously have not read the entire internet, I do have yet to encounter a bootcamp whose marketing strategy ISN'T centered around the promise of becoming a full-time developer upon matriculation.

Programming skill is a hiring desirability multiplier because the context of the work is usually just as important as the work itself.

Eg. Accountant -> Code Camp -> Tech Consulting -> CTO for accounting firm.

Reactor Core, the parent company of Hack Reactor, recently proposed a standardized methodology for the industry.

http://www.prnewswire.com/news-releases/reactor-core-announc...

Where's the link to the supposedly methodology? That's a pretty crappy press release...

They will teach you that methodology in their bootcamp. Sign up today!

The first step is free, just for signing up for the email list?

[I'm a founder.] It was on our site [1] until yesterday -- apparently it is now hidden behind an email collection mechanism. Here's the hotlink[2].

[1] http://reactorcore.com/student-outcomes [2] http://reactorcore.com/wp-content/uploads/2016/05/StandardSt...

I agree. I originally wrote this article "Are dev bootcamps a scam? A hacker's perspective."[0] which actually caused quite a bit of drama when I then had dev schools up in arms about opening up the books and proving their success/placement rates, etc.

Then I decided there should be a better way to find reliable, quality bootcamps so I built a review site for Dev Bootcamps.[1] The response has been great and what was once reviled by dev schools is now seen as a source of distribution and qualification for validation because we help so many people decide where to go.

[0] https://www.techendo.com/posts/are-dev-bootcamps-a-scam-a-ha... [1] http://schools.techendo.com/

https://schools.techendo.com certificate is expired

Yep. In the middle of taking care of that now. Thank you for the heads up.

I agree with the author that bootcamp "statistics" should have some kind of standard (that bootcamps are free to use or disregard, but that way applicants will know the methodology being used is either the published standard or one that is not published). My one quibble is with his claim that no one goes to these schools to become product managers - plenty do. There are enough PM jobs that require software development experience (reasonably - the best PM's I've worked with could all code well enough to bang out a prototype or proof of concept) to justify learning to program. Outside of that, couldn't agee more.

Isn't the problem along with these somewhat dubious "Hacker Schools" or "Bootcamps" the demand side for a "shortcut" to an instant developer title or less likely an instant $100K/year job?

The reality is that there are no shortcuts. What ever happened to the good old fashioned "take an entry-level support position and work your way up"? That gets you both a foot in the door and get real rubber on the road experience? Is it sexy? No. Does it work? Yes.

I feel like it's very easy to fudge the numbers just due to all the different factors involved. Would student reviews be a better way to evaluate bootcamps? I know I've seen some bootcamps on yelp with reviews. This seems like it would be better than a self-reported statistic.

In California, the Bureau of Private PostSecondary education is coming to implement transparency standards in a big way. In the next year or two, you will see several bootcamps forced to be more transparent via approval to operate, or be forced to shut.

Went to Maker's Academy London, lots of flavour of the week development languages,corporate cult vibes, and too expensive for what it was.

General Assembly is working towards just such an approach:

https://generalassemb.ly/blog/measuring-what-matters/

Note: Most of the interesting stuff there is in a PDF.

(Also, I work for GA, but am not speaking on behalf of the company.)

The PDF is interesting, but reports no actual data from GA... Seems like that would be relevant, if they are trying to push for more transparency!

Agreed - why make a big deal out of publishing a "framework" and then not release any data to go along with it?

So tired of bootcamps. I went. It sucked. Don't go.

Many people that attend code schools have no interest in getting a software engineering job. I think asking code schools to base their metrics on employment rates would have bad side effects.

This would cause schools to reject applicants that are interested in learning coding for alternatives reasons. These reasons range from Mom's wanting to learn so they could teach their children, wantrepreneur's wanting to build their MVP on their own, or just someone that is happy with their career path, but wants to take a break to learn coding for fun.

Do you think Mom's wanting to teach their children coding should be rejected from code schools, because they don't want to get a job and thus would hurt their numbers?

Do you really think that people spend tens of thousands of dollars and countless hours (away from family, might I add) learning to code just so they can teach their children?

Maybe these deserve different kinds of code schools?

What are you talking about? Where is your source that people that attend code schools don't want jobs?

I also think this comment misses the point of the article. The author is not proposing that bootcamps reject people based on their career objectives. He's pushing for greater transparency in reporting and urging other bootcamps to publish more than just the splashy stats displayed for marketing purposes. People make huge financial and emotional investments to attend a bootcamp; the author argues that bootcamps should do better by their prospective students by helping them to understand just what they're purchasing.

Reply on news.ycombinator.com