I charge 100$ an hour to do coding challenges. Recruiters guffaw when I tell them (of course they do), but I don't do the ol' jazz hands routine with the money makers for free. Really, the positions they're seeking to fill pay > 100k. What's a few hundred to a candidate if you really want them to demonstrate a skill? Pay that and I don't care if I'm actually closing a real ticket for you as part of the interview process.
Once I wavered on that for a data science position. They had me do some predictive analytics on a dataset. I've built a pipeline for problems like that so I figured I'd spend the 20 minutes spinning it up... I ended up spending an hour creating a rather detailed writeup about methods used and my analysis. The stake holder got back to me asking for changes to minor format things with my output data. (headings, serialization of booleans.. minor shit like that) I came to the cynical conclusion they were getting free data science work and the position was an illusion. I asked if that was the case and never got a response.
Quite a few. I've been a full time self employed consultant for 3 years now. Yes, I'm being cute -- but the distinction between customers and recruiters who trick me into answering the phone by spoofing my area code is irrelevant to where I draw my lines. I like the internal consistency in my handling of either situation.
> Employers already have most of the power in this relationship
I feel like it's the exact opposite.
I once was at a team where a coworker left. He was one of two developers working on a subsystem. The subsystem was highly valued by the company because it would result in faster payments and less labor in processing payments. Even though they replaced him with another developer the project still took 1.75x longer to complete compared to the previous pace.
When a developer replaces a company they continue to pay the same green money. When a company replaces a developer it can take months or years to retrain him with institutional knowledge. The asymmetry gives developers more power in the relationship.
We sometimes hire mid/jr-level FE developers, which means we get _lots_ of bootcamp applicants.
It feels like the main goal of the bootcamp is to produce portfolios/resumes that make you indistinguishable on paper from developers that learned any other way.
Not necessarily bad, but the quality difference among bootcamp grads (from the same program!) is crazy. Some people understand the fundamentals of what they learned and continue to learn and expand their skills, but some are just copy/pasting code and debugging by typing random character combinations until something works. Both got through the bootcamp with identical group-project portfolios and class assignment personal websites and resumes.
We've recently started sending an at-home coding challenge to all jr. FE applicants, just to cut down on amount of time wasted if we bring them in and they don't know anything. This has worked okay so far.
I'm generally very cognizant of wasting peoples time, so I don't want to do this as a general rule for more experienced candidates. But for junior/first-job candidates, I'm not sure it should be as offensive. Do any of y'all see a reasonable distinction here?
Is a bootcamp grad much different than a university grad?
While in school, I read blogs claiming that the majority of CS grads can't code. I got the hint this may be true in one of my 3rd-year classes where we had to work in groups of 2 or 3 over the course of the semester to write a program that was essentially like an alpha version of SimCity. At the last day of class, we had to present our work. Half the groups didn't complete the project, and one group only had a blank window with a Help button that presented a wall of text. Then, during my senior project, in my group of six, two of them admitted to not knowing how to code, with one of them saying they didn't even want to code, they just became a CS major because they heard CS majors have low unemployment and good salaries.
I've digressed a bit, but...
When I finished my degree, I didn't have a software engineering internship under my belt, so I made sure to list personal projects on my resume. I think the presence of projects done on my own time for my own enjoyment and learning was a key factor in landing my first job.
So no, I don't think there's anything offensive about rejecting a candidate if a bootcamp is all they have to show.
I’ve worked with people who have MBAs but aren’t really developers or even devops types.
It’s interesting to see them get frustrated so easily and have a hard time plowing through problem after problem making small changes until they finally get to something that works.
Someone who has only done a boot camp hasn’t demonstrated that they can work on problems for weeks or months before they can see real progress.
Someone who has an actual degree from a legit university or college, has at least proven that they can spend months or even years working on something until they get their reward. They won’t necessarily be able to program their way out of a wet paper bag, but they have at least one advantage on people who only have the boot camp certificate.
I've been hiring people for a long time and have always been against coding challenges. That was possible when I was only looking for more senior roles or hiring junior roles at a startup or small company. I'd generally talk to anyone that didn't appear incompetent to HR during an initial screen.
Now that I'm hiring for multiple junior openings at a fairly well known public company, hundreds of resumes come in each week, and I added a code test is there to help weed out those who are tragically lacking in skill or effort. The test is about as simple as could be and I generally don't even look at the score. I do look at the code, however.
Offensive, interesting. Yes, I don't think it's offensive. I do think the attitude of "my time is worth more than yours" isn't one I particularly appreciate. When I was interviewing and got at-home coding challenges, I appreciated it when the assumption was I was going to get actual helpful feedback if I didn't pass to the next level. A solid code review, if you will.
At least then there's some sort of reciprocity in time.
I am moving in this direction after having taken a few coding challenges and not heard anything back from the company. If I am going to spend time on a coding challenge, I am expecting at least an acknowledgment that I am no longer being considered for the position. Coding challenges are a time commitment and if a company can't even be bothered to commit to at least being polite about it, then it's a clear indication of how they view employees.
TBH, for the positions I applied to, a coding challenge came _after_ a face-to-face interview. From what I later learned, I have been mostly given low-priority tasks built upon the actual codebase, but I was expecting that much at that point in the process (why bother spinning up a googol+1st blog implementation?).
Having just been through the interview process, i ended up having to go through a few coding challenges. Luckily for me, i never got “ghosted” after sending one back.
The important thing to me is that the coding challenges have a reasonable scope. I really appreciated two of the companies that gave me an already set up project (it was not their actual codebase) and then required me to build an addition to it. That feels like an accurate way to test someone’s skills.
As someone who doesnt excel at some of the “leetcode” interview questions as much. I have found challenges to be a welcome alternative when done right.
I did experience some companies who would give a project that was wayyy too complex and they estimated it would only take one hour. Those are the places i would reject without doing it. With the market as competitive as it is, you need to have some limits with these things.
As an interviewer, I try to ask a consistent question in a consistent way, so that I can calibrate my feedback against all of the people I've interviewed before. It's nice that you claim to have some previous code, but most of my candidates aren't able to share their code, and some hopefully small number of my candidates are going to submit someone else's code and claim it as their own. I have worked with and interviewed people that can discuss all the things, and run through other people's code as if it's their own, but are unable to write their own code.
That said, exceptional candidates are exceptions and maybe should be handled outside the normal process.
THIS. I have 8 years of open source work in bugzilla and github. If companies really give a shit, they would review code you already wrote and ask questions about your body of work. Not some arbitrary test.
Asking to speak to an engineer isn't going to change much. If you really mean business insist they pay a nominal fee for your time. Not an hourly contract rate, more like the hourly salary rate of a mid-level dev. It's peanuts to them so if they won't pay they don't value your input so it's game over.
I agree, but it's probably no a thing the recruiter can do without going through the boss of the boss and hassle with accounting department. Until a big body of of candidates require it, nothing will happen, and if too many require it, the employer will probably just quit giving home work.
This concept of whiteboard interviews and assignments is quite alien to me. I've got three different software jobs since graduation and not one did any sorts of test they just trusted that I would not apply if I couldn't code. Never heard of any colleagues or former class mates that done any either.
sorry but this is as absurd as saying don’t let line cutters into traffic. yes, “we”
should band together to end the practice, but the fact that “everyone else” is going to do it means it will continue.
rather than address the call to action to candidates, how about asking folks to refuse to give a coding test?
Once I wavered on that for a data science position. They had me do some predictive analytics on a dataset. I've built a pipeline for problems like that so I figured I'd spend the 20 minutes spinning it up... I ended up spending an hour creating a rather detailed writeup about methods used and my analysis. The stake holder got back to me asking for changes to minor format things with my output data. (headings, serialization of booleans.. minor shit like that) I came to the cynical conclusion they were getting free data science work and the position was an illusion. I asked if that was the case and never got a response.
Learned my lesson I guess.