161–170 of 1001 posts
The bottom line is you can't instantly validate how someone will perform over the course of a year or more. It takes time. But nobody wants to spend time so...
As a developer and a manger, there are many candidates who fake or bolster a resume, especially in tech. Additionally, there are some developers who are incredibly smart, but take 2-3x the time as someone who is not as senior to do the same task. The only way I've found to reliably hire is a coding test. I think it's ethically irresponsible to choose actual client work for the task. But it is important to create a te…
What about paying someone to do client work as a test? That is what I've found to be the best approach for me. It gives the candidate a very concrete understanding of the kind of work they'll be doing as opposed to a more contrived example. It comes at the cost of paying for poor work but usually some portion of it is salvageable.
My tests took about 10min to write and 1-2hrs to complete and then we could do an apples to apples comparison across candidates.
My concern about client work would be the time it would take to bring someone up to speed as well.
If you're going to expect people to do real work, you need to pay them. This is common in a field like journalism. Want to do a one or two day trial of a reporter? Pay them to write for you! Have them work in a real environment and do real work for publication. If you're not willing to pay people for their time, and it's not an executive position, asking for days of a person's time is asinine. Also, if the homework i…
Exactly. Just a week or two of contract work will tell you if the candidate is a fit or not. A 30 minute phone screen will tell you if a candidate is worthy of a contract. I think companies don't do this because a. They don't want to pay and b. They don't get to gloat 'we have a super rigorous hiring process'
This is still far cheaper than the costs of hiring the wrong candidate.
I actually prefer take home. I’m not great at white board solutions as they’re basically asking you to solve something immediately with a gun to your head in real time. It’s very unnatural. Take home gives me time to think about something more like I would job. That said there needs to be a balance where the tasks are enough of a test to give indication of quality of work without taking 3 full time days to finish.
A sixty minute screen share is the sweet spot for me. They get to use whatever environment they are familiar with. Seeing someone work in their environment also tells you potentially a lot about the candidate.
Staff engineer that can't navigate their debugger?
Senior engineer that codes by copying every line off of stack overflow and mashing the keyboard until the lines somehow work?
Bootcamp graduate that builds test cases in Excel?
Junior engineer that writes code in notepad.txt and pastes it into a Python shell?
All things that I've seen happen that tell a different story than a whiteboard or a 3 hour submitted homework test.
Nowadays, I reject more companies than I interview for based on an initial phone conversation. I have requirements in what I look for in a company and position, and some simple questions help me resolve many of the primary requirements. Given a take home exam, I will have invested hours of my time, before given a chance to screen the company.
Just say no. End the madness.
EDIT: I should add that when I mention a phone conversation, I meant with someone on the dev team, not HR. A take home exam after only an HR phone screen is not acceptable.
I prefer homework to timed, monitored programming sessions. But the task should never be a big task. When i interviewed for a backend development position, the task was to set up flask with two routes, connect to a dB and retrieve a couple of rows when user visited the routes. It was not hard even though I never worked with flask before, and it wasn't supposed to be hard either. It was a test if I had basic programmi…
The worst one was for Optimizely, who asked me to write an entire website (front-end using React, backend in Python) which will display a list of products from a database, including full tests. I determined that boilerplate and configuration would take me about 5 hours, with maybe another 5-6 hours of actual coding. They even offered to pay me something like $15/hour (literally nothing) to do this. I said no.
Earlier quoted context omitted.
Exactly. Just a week or two of contract work will tell you if the candidate is a fit or not. A 30 minute phone screen will tell you if a candidate is worthy of a contract. I think companies don't do this because a. They don't want to pay and b. They don't get to gloat 'we have a super rigorous hiring process'
How many engineers would accept an offer which is conditional on "a week or two of contract work"? Most of the engineers that I've hired came from another job (not all though), and so they need a confirmed offer before they are willing to jump ship. So this practice introduces its own selection bias; you'd only be sampling from the pool of currently un-/fun-employed, and those who hate their current job enough to qui…
Earlier quoted context omitted.
How do you try to hire people who currently have jobs with strict anti moonlighting policies? (Or maybe the answer is just don't bother, and you can afford shrinking the applicant pool without any big adverse effects)
Has anyone tried offering to pay but permitting candidates to decline payment? (My current and previous jobs have these restrictions, but they also pay well enough that I care quite a bit less than I did when I was a college student / funemployed / etc.) It seems like it's likely to work similarly to bug bounties: it's not like you stop getting security reports from people who can't accept the bounties, you just don'…