depends on how involved in the domain you’re going to be

also as a tech person you’ll usually be expected to learn the domain stuff on the job

I’d be careful with that if you don’t want to end up as a product manager for example

I was asked to try to compare salaries for Big Data engineers in London knowing Python/Scala and Clojure. The hypothesis is that in Big Data space Python and Scala are standard therefore most people would know at least one of them and requiring Clojure on top would make them more expensive. Any ideas how people would approach such problem? So far my meagre list of ideas: 1) asking friends 2) looking at advertised jobs on job boards 3) https://www.itjobswatch.co.uk/ 4) asking here! I don’t think there is a good answer, but if anyone has any I would be grateful!

the sample size you’re going to get matching the requirements will be tiny and have a huge variability in experience

yup, exactly!

I guess this is more general question then Big Data, Clojure and London

How do employers decide on salary bands given really limited information of the general market?

How do hiring managers within the company negotiate the renumeration for specialist positions?

Apparently this is useful: https://www.payscale.com/

@dev964 ^

But I would also be interested in knowing how others figure out what salaries are “standard” and “fair”!

from employee perspective or employer perspective?

I for one don’t think that such difference between software engineers and everyone else exists

Both perspectives are useful

Whichever side you’re on, it helps to understand the other one

Personally I usually negotiate as high as I can get away and then some more

When working the salary knowledge would spread in one way or another

and sometimes that means others would ask for higher salaries

One way that my startup determined salaries to offer was based on information provided by VC networks. A few VCs have private networks for their portfolio companies that maintains things like comp data and they have a recruiting team and you can ask questions to other members of the network Quora-style. Usually the comp data is broken down into equity and salary and stage of the company. Obviously people negotiate and positions dont always neatly fit into the data but it was at least a starting point.

i think in a negotiation it's probably worse for a company to low-ball than for a potential employee to high-ball. If you low-ball someone by too much they'll be offended and then your negotiation is off to a bad start or they'll just walk away. I have a hard time imagining a company, once you've gone through the interview process and the team likes you, getting too mad at you for asking for more than they're willing to pay. The exception might be for very large companies that have a lot of interest from devs to work there.

Yeah, worst they can say is that they're not willing to go that high. Then you have to decide if you want to go low enough to meet their offer or not

here’s a decent article on salary negotiation: https://www.kalzumeus.com/2012/01/23/salary-negotiation/

Incidentally, they linked this article over on #jobs https://circleci.com/blog/how-we-interview-engineers-at-circleci/

It blows my mind.

Out of any group of 1000 applicants: 250 will pass the initial screen. 117 will pass the hiring manager phone screen. 44 will pass the micro-skills assessment portion of the interview. 7 will pass the macro-skills assessment. Fewer than 3 will pass the on-site interview and receive offers.

Those numbers don't seem to jibe with what they write at the bottom of the article, > For example, right now with an average .5% conversion rate for applicants, we’d need to look at up to 4,000 candidates to hire another 20 engineers. so they're hiring more like 5 per 1000 than "fewer than 3", if it was 1 or 2 they would need to look at more like 10-20k candidates for 20.

5 per 1000 is still a damn small number

even then i wonder if they're not exaggerating a little.

It's easy to get "applicants"

i mean they can probably just throw the vast majority of the applications in the trash because they're probably not even applying for the right job, or forgot their resume, or are people trying to reach customer support through the job application site (seriously)

O_O

"hiring manager phone screen" -> hires of 1% is too low. That's a waste of an actual hiring managers time

the start number and the final number might be right (0.5) but I think the numbers in the middle are probably not what they're saying there

what's attracting so many developers to work at circleci. Compensation or coolness factor?

How representative is that for a hiring process anyway?

im not sure what their compensation is like...but they seem to offer remote work, they're clojure-based, and the software is pretty nice

I remember reading discussions where hiring managers would talk about how a lot of candidates can’t code fizz buzz

but I’ve always assumed this was for entry roles

where did 1000 clojure applicants come from? Mars?

haha that's probably why they can eliminate so many of them

they hire remotely worldwide AFAIK

yeah there are definitely a lot of people out there who can't code worth a damn

but they are still able to get jobs, they just work for SME's doing boring stuff in C# or Java. Nothing wrong with that.

a lot of organisations that kill your soul …

I keep being surprised how bad software organisations are

so one time this company had me do an online interview screen and they asked me to do FizzBuzz. The nice thing was that you could pick a language to do it in, so I chose C and did fizzbuzz using bitwise operations and no if / case /ternary statements. I never even got a call back! Come on...

sorry, i lied, it did use a case statement

ooh

https://aphyr.com/posts/340-reversing-the-technical-interview 😄

i guess they didnt feel like hiring smartasses

this article is pretty good haha

check out the next two as well, it’s amazing if you like smartassery for fun and profit

Job interview process is a piece of information about the company in itself

if you pass you’ll join the team that was selected based on that process and even likely designed it

or had input into designing it

there’s definitely always a number of irrelevant applicants

I once posted on a Who’s Hiring thread on HN for a backend clojure position

and got a bunch of canned emails from devops contractors, probably scraping the comments for email addresses

haha

😞

@jgh I guess it depends what they're looking for. Here we value code clarity and 'easy of understanding' for someone new coming onto a project. Especially when working in less common languages like Clojure over, say Java or JavaScript. So as long as there is no significant and relevant performance gain, writing something by being 'clever' using bitwise operators instead of the normal/simple case would actually count against you

Maybe unless you provided a readme that said "I did this because it demonstrated an X% performance increase over if/case statements as tested using Y"

yeah i know, let's just say i wasn't too broken up about not getting a call back 😉

https://www.reddit.com/r/cscareerquestions/comments/95dgrx/i_am_absolutely_mortified_and_embarrassed_beyond/ tl;tr Guy killed owner dog during interview, by accident.

oh man

Regarding the conversation in #jobs about Toptal I worked with a client who hired a very talented Clojure(Script) dev from Toptal. At least one person has gotten Clojure work off of their site.

now do you take that job if you get a call back? That would be a really awkward position to be in. Thankfully it doesnt appear to be their fault, at least from how they tell the story.

I probably wouldn’t unless if it’s my dream job. Even if it’s no one’s fault, it’s just going to be a cloud over your head

and you never know how people might be subconsciously biased

much better to go somewhere where you start with a mostly clean slate

@jsa-aerial once you’re forced to use jenkins for a while then Circle CI is very cool

I'm the Technical Recuiter at CircleCI. Feel free to DM me with any questions