Once recruiters figured out that placing these under qualified people into jobs was not a good thing, the bootcamps started to train people how to fake their resumes. Just like how people figured out that if they want a specific job, they only need to update their resume to include all the keywords in the job description.
When I was hiring and screening hundreds of people, applications would show 'projects' completed on them as if they were real work. Hired.com was full of this for a while until they started blocking them there too. People would go so far as to make a domain name for a class project appear as if it was a real company. You'd go look at their github and realize they only made a few commits towards the 'project'. How is that a demonstration of skill?
I'm sure a few people have come out of these bootcamps with some real knowledge and skills. But this is the edge case, not the norm. You can't shortcut your way to a well paying software engineering career by spending a ton of money.
Folks gotta start somewhere. If someone is sufficiently capable and motivated to put together a basic app why not count that experience? Surely if they built it then their work is exactly a demonstration of their skill.
It was a 'team' effort. So, when you go look at the git history of the 'project', you can see who in the team actually contributed. Usually, it wasn't the person looking for a job. Then you wonder, ok, maybe shared computer... which makes it even harder to figure out. How is that a win for people who are paying tons of money to get an 'education'?
I believe strongly that software engineering is an art form that requires tons of practice to get good at. As such, you wouldn't expect a designer or artist to show up at an interview and not be able to show examples of their work. Just like a software engineer should be able to show something as well. Of course, the counter argument to that is maybe they didn't have time to work on something extra curricular. Except for the fact that the people who did the extra curricular work in high school, got into better colleges. Go figure.
Even worse is that I did try to interview some of these people and it became clear that the bootcamps were about training for the minimal viable education. Leaving a lot of basic fundamentals out. I'd ask what idempotency is and get blank stares. So sad.
Yes, you have to start somewhere, but that doesn't mean you are immediately prepared to get a job as soon as you graduate. That is part of the broken promise of these bootcamps.
Idempotency is a good example of a trivia term that would be unreasonable to expect most recent CS graduates to know. Not to say that it's not a valuable property for a service or library or whatever to have under certain circumstances-- but it's not a reasonable way to decide whether a candidate for a junior position is a good fit.
Huh, I did a diploma in computing (UK) in my spare time about 12-16 years ago and remember covering it in a databases module (and possibly when doing OOP with Smalltalk?) - I'd go with "it's about repeatability, knowing that sending the same request, eg SQL query, will return the same results".
That's a little off the mark, as I've checked, but I'd think a full-time CS (as opposed to programming or computing or IT) student would be able to define it very readily ...?
If they submit projects they did not contribute to then that is an issue. Your other points however, I disagree with.
- all code I write commercially will have dozens of contributors. It's a team sport and you need to be able to play well with others to succeed.
- the best place to get experience is in the works place. That is why folks hire juniors. They will have massive gaps in their knowledge, but they will also be keen and in learning mode. Help them improve and become great developers. They will do much better in a commercial environment with good guidance than struggling at home alone
I wasn't arguing against working with other people. LOL.
What I'm saying is that you should have clearly identifiable portions that you can point at as something you did. I also think that it takes extra curricular activity to truly master a subject. You can't just rely on others to show you everything. You have to experiment and learn on your own too.
>>When I was hiring and screening hundreds of people, applications would show 'projects' completed on them as if they were real work.
You have to start somewhere, right? Those projects may not be "real" work but they may be enough to have given the candidate exposure to various problem domains and experience in solving problems using code, which should really be sufficient if you are hiring entry-level devs.
I mean what is the difference between the projects I do on my weekends and the projects a code school student completes as part of their curriculum? At the end of the day there is only one thing that matters, and that is whether the candidate you are interviewing has the necessary skills and insights or the potential to gain them.
Keep in mind that many computer science graduates can't even solve fizzbuzz.
"Keep in mind that many computer science graduates can't even solve fizzbuzz."
I have encountered far more people who repeat this "fact" than I have encountered CS graduates who actually cannot solve fizzbuzz.
Closer to the truth is the statement that many computer science graduates can't solve a topcoder-elite problem, on a whiteboard, within 45 minutes, while talking out loud the entire time. Because that's closer to a real-world "fizzbuzz test" than anything I've seen on Joel Spolsky's blog. (sadly)
The difference between the projects you do on your weekends and projects a code school student completes are absolutely immense.
An applicant with strong weekend projects implies:
- They have a passion for building software
- They are self directed learners
- They have the initiative and determination to build things and ship them
- They managed to navigate problems without having an instructor hold their hand
- They actually did the work, and aren't taking credit for a group effort they may have had minimal impact on
Code camp projects don't say much of anything either way to me on a resume. They don't rule out the above but they don't support it either. They basically say:
- I care enough about learning to code that I paid for a code camp
- I didn't get kicked out or completely fail out of the code camp
- I realize that employers value seeing projects on resumes
To me putting the code camp projects on a resume when they are not clearly marked as being part of that code camp is as dishonest as doing it for college projects when they are not listed under your education background. I've seen countless resumes come in where the projects from code camps are presented in a way so that the recruiter/hiring manager thinks they were self directed projects the person did on their own. The code camps know that this is a strong signal for top talent, and so they mentor the students to craft their resumes to mask the fact that the projects were assigned work.
The result of this, sadly, was that enrollment in code bootcamps became a first order screen for me. I'm sure there were some good people in there. But the work required to actually decipher what the resume was actually telling me required too much effort. (It would require opening the projects they did, trying to figure out if it was assigned or not, if it was a group project, and what contributions they made directly.) In the end, this combined with the fact that many code boot camp applicants are woefully underqualified made it not worth the time to sift through them to find the best ones.
As I said above, 'group effort' doesn't mean the person actually did the work or learned anything really. A project YOU do on the side means that YOU did it. =)
Once recruiters figured out that placing these under qualified people into jobs was not a good thing, the bootcamps started to train people how to fake their resumes. Just like how people figured out that if they want a specific job, they only need to update their resume to include all the keywords in the job description.
When I was hiring and screening hundreds of people, applications would show 'projects' completed on them as if they were real work. Hired.com was full of this for a while until they started blocking them there too. People would go so far as to make a domain name for a class project appear as if it was a real company. You'd go look at their github and realize they only made a few commits towards the 'project'. How is that a demonstration of skill?
I'm sure a few people have come out of these bootcamps with some real knowledge and skills. But this is the edge case, not the norm. You can't shortcut your way to a well paying software engineering career by spending a ton of money.