So are they looking for C OR C++ programmers, either may apply? or are they looking for strong understanding of the nuances of both C and C++?
I encounter a lot of "C/C++" on resumes and after interviewing a lot of folks, I've grown a distaste for this joined tag. The problem is that when I see "C/C++", I never know until I interview a candidate whether the intent is that "C/C++" is...
1) Shorthand for I program C and I program C++ (I know both well)
2) I program C with classes. I don't really know OOP, but I fake it with big class buckets.
3) I program in C++ and know OOP well, but I would get lost in the strict procedural world of C.
If you don't get where I'm coming from, I've programmed in both for many years, and I will simply say that "C and C++ are two paradigms separated by a common language."
I won't speak for Slava, but I think generally the idea is that knowing either one well enough implies a certain amount of knowledge about computer guts, which in turn implies that it shouldn't be too hard for you to learn the other.
Plus it's not entirely unheard of to do the core of the application in C, and offer a C++ API to users. There is a good reason to have both expertise inhouse.
I find the best programmers know C, C++, C#, and Objective-C. They know the pros and pitfalls of each language:
1) Procedural vs OO
2) Manual vs automatic memory management
3) Naked pointers vs smart pointers vs garbage collection vs reference counting
4) Method calling vs message passing
5) C libraries vs C++ libraries vs .NET frameworks vs Cocoa frameworks
6) Zero runtime vs thin runtime vs rich runtime
I am impressed by the inclusion of the salary chart right there in the job posting, as an engineer who dislikes the salary negotiation game (I am inevitably at a strategic disadvantage, playing against managers who have far more experience at the game than I).
At least this way, the engineer knows where he/she stands.
Not only that, they cut straight to the chase and put the value in % of the company.
In fact, has anyone ever seen a job post like this? I know I haven't; I'd remember. It's highly unusual in its straightforwardness. I'd guess that's pretty appealing to the kind of hacker they're looking for.
There aren't any others that I know of. Most people are incredibly secretive about their stock options - to the point that they give you a vague absolute number (that will likely split or merge at exit) without disclosing a percentage. It's mathematically impossible to evaluate offers like that. "Ten thousand shares" sounds better than 0.001%, though.
There are some very good arguments for keeping the information closed, but we thought we'd try it. So far we've had very positive reaction, but the real test is internal dynamics as the company grows.
Just out of curiosity, what are the arguments that you considered for keeping the information closed?
A somewhat related anecdote... I recall with chagrin the first startup I joined. The CEO was a friend of mine. I kept asking him how much equity he was offering as a percentage. He kept responding evasively by snowing me with all sorts of dealy jargon that I didn't understand. In retrospect, of course, he was concealing the fact that the percentage was negligible. At the time, though, insecure in my knowledge of business, I walked away thinking that I was a pathetic noob for having the idea that percentage was a meaningful metric. (This guy was good at what he did, which was deceive smart but gullible people.) That experience is one reason I like your job post. I totally agree with the commenter who said that they hate the jockeying of the negotiation process and would find it a relief to just know up front.
Anyway, I find your post to be an interesting experiment and I think many of us would like to see a follow-up report on it someday.
Suppose you pay your top guys $100k, and a candidate comes along with about the same level of skill and experience that you really want to hire. But he won't join for less than $120k. At this point you either have to pass, or you have to bump everyone else's salary. If you have ten guys, instead of paying extra $20k, you suddenly have to pay extra $220k - a massive difference. Same goes for stock options.
I haven't heard a good argument in favor of hiding stock option percentages - I can't think of any other than concealing the fact that your share is negligible.
"Introduction to Algorithms by Cormen and Leiserson is your trusty friend."
I have this in my bookshelf. Every time I walk by, I can feel it gloating at me with a smug expression. The rare times I make eye-contact with it, it seems to say "Want some more? Come on back, I've got more, I'll teach you another lesson!"
This book is like toothpaste - no matter how long you use it for, you can always squeeze more out of it. There should be a theorem of some kind about that.
I think I might be somewhat representative of the target audience for interns (maybe), so here's my two cents
- Really cool stuff. Anything heavily concurrent in C is just cool.
Highlighting that even more is going to get you more of the right
kinds of candidates IMHO.
- Call for interns in the fall. Offers start going out for
internships in the late fall and early winter.
- The world domination stuff is hokey. Every company is feeding
us this stuff.
- Do interns get free housing? What are the perks?
- Your pay for interns is very standard (which is a good thing).
Thanks for your feedback, we'll incorporate it into the jobs page. To answer some points here, we do not arrange housing for the interns (you can probably crash our place until you find a sublet, and we can put interns in contact with each other so they can self-organize). We didn't completely define the perks yet, but it'll be pretty standard stuff.
I agree that the world domination stuff is hokey, but it's catchy and it grabs people's attention, so it serves its purpose pretty well.
If in doubt, apply! We're pretty flexible when it comes to the details.
This is the kind of job description that I wish someone in India would advertise. The job boards hereabouts are full of the outsourced "maintain legacy enterprise systems" crap jobs.
No I am not looking for a job [1], but this ad crystallized what I miss in the Indian sw/startup ecosystem. Too many people building db backed websites - clones of successful western products. Not much hardcore tech.
Devs in the USA are lucky to a degree they aren't aware of.
(Mildly) depressed.
And as for jobs with upfront info about salaries/equity, we can only dream.
[1] I am working on a cutting edge (non startup) project that involves a lot of Machine Learning and algorithmic complexity. If I were to advertise I could mostly cut and paste this ad - replace DB internals with Machine Learning - but this is very, very rare in India and I know a lot of talented devs who would give a great deal to be able to apply for such jobs without going through visa hassles and so on.
"Too many people building db backed websites - clones of successful western products. Not much hardcore tech."
Isn't Twitter a DB backed website? What about Facebook? Try to understand that there is a need for various types of web-apps in the world. Everyone doesn't need to build MySQL storage-engines for SSDs, for example.
Don't underestimate the effort that goes into making a db-backed website available to a few million people on the Internet.
Recently I put a cutting-edge, hardcore (low-level, C, cpu architecture, smp, concurrency type job) startup job post to Naukri and I only seem to get mediocre responses.
I believe it's a two sided issue, there aren't that many talented people or as the saying goes, those that are talented have already moved to western countries.
I really really would like to find talented people from India but it doesn't seem that easy.
In my opinion, the current job sites in India are completely broken for the smart programmers who are looking for interesting work. And the few companies/startups who would like to hire such people again fail miserably (most of the times) when they try out these sites.
These sites cater to the majority - companies doing low-end services work & average programmers. (Sorry for sounding elitist here.)
It is a very good opportunity for a startup to bridge this gap. Make a smart programmer meet a smart company.
HN seems to be a good place to reach out for good people but a country/location specific (apart from Silicon Valley) job post doesn't stand much chance of being on the front page for long.
It would have been great if we had a job board for HNers where job posts can only be made by people with a certain karma threshold (e.g. 500) and each of it costs 50 points for the submitter. Allow browsing through the job posts based on locations/countries and type of work.
It is no different from any other job listing out there, but since it comes from a HN user I think it would have a higher chance of being interesting enough for people who hang around here (and other smart people as well who are not on HN).
I've never needed to use job sites to get a job, touch wood. I am not using any. My point was that job sites are deluged by enterprise sw jobs.
Note that the job post under discussion wasn't put up on Monster.com. How then is your experience equivalent?
"Recently I put a cutting-edge, hardcore (low-level, C, cpu architecture, smp, concurrency type job) startup job post to Naukri and I only seem to get mediocre responses."
Naukri sucks. And advertising on Naukri rarely works unless you are some kind of bodyshopper like TCS or Wipro. No self respecting hacker looks ofr a job on Naukri (or any job borad for that matter).
" there aren't that many talented people"
yeah sure! Indian developers are untalented because they choose t live in India ;-)
" those that are talented have already moved to western countries."
partially but not completely true. Many ultra talented devs don't want to immigrate. Just shows you are fairly unplugged from the local hacker scene.
"I really really would like to find talented people from India but it doesn't seem that easy."
Well it is not as easy as putting up a job posting on a sucky site(naukri) and sitting back and waiting, which is what you seem to have done (I am extrapolating from your posting here, I could be wrong).
More constructively, do you have a website for your company? Are you funded? Do you have a blog? Who are you ? What exactly are you doing? Are you plugged in to the local hacker networks? Are you hacker yourself? Do you have any open source contributions we can judge you by?
If you can provide a link to your job post I may be able to give better feedback. My email is in my profile as well.
would you mind providing your contact information (didn't seem to find anything in your profile). i am in india, with approx the skill-set you are looking for, and am on the fence w.r.t the job scene (our company just got bought by a large networking company)
The problem is, in the US even jobs where you will actually be maintaining legacy enterprise systems are advertised like this these days. Having seen so many job ads in this style recently, some cooked up as job-bait by recruiters, I tend to trust ads with more grounded language.
We really are doing what the ad says. It's unfortunate that many people use bait-and-switch. We specifically tried to choose the wording to be as forthcoming as possible.
I believe you - I was just addressing plinkplonk's broader point about ads in this style and offering my own recent experience in the job market as a counterexample.
Sure, but this is a YC funded startup (iirc) and I have a fair idea what they are doing and I am fairly certain that the ad describes what they actually do/want. What is wrong with the Indian sw "scene" is the almost complete absence of such companies .
Who cares about professional recruiters pitches anyway? :-) .Most of them are liars and cheats (or so it is safer to assume, at least hereabouts), but yes, fair point!
The job boards hereabouts are full of the outsourced "maintain legacy enterprise systems" crap jobs.
As someone who has both done Big Freaking Enterprise for the last three years and managed outsourcing of the same (to India, actually): please, find a way to fill my customer's needs in a manner which doesn't require custom development and put us all out of work. I think there are many, many opportunities for eating the enterprise elephant from within. (You also have to be able to sell them, obviously.)
" I think there are many, many opportunities for eating the enterprise elephant from within. "
Yes I agree. I guess systems can be lucrative and soul killing at the same time.
I worked in enterprise dev for many years, last at ThoughtWorks (which is a fine company in that space - If I ever had to outsource enterprise software, I'd probably use TW (being somewhat selective about team composition) or a team of ex TW ers. I would never outsource to Wipro Infosys et al).
I grew bored of enterprise dev and dropped it completely almost 8 years ago.
Enterprise systems are interesting when you own the enterprise! When you have to work on outsourced crap codebases, life sucks.
Yes I agree. I guess systems can be lucrative and soul killing at the same time.
It's probably not a coincidence that soul killing things are also lucrative.
I'm wondering though why there aren't more people eating at the elephant. Replacing broken systems with new, non-broken ones doesn't strike me as soul killing, and it's clearly potentially very lucrative.
Machine learning tends to need specialized datastore infrastructure as well (look at DirectedEdge or the full text search indices). I'm curious, what is your project about?
I am working on a couple of projects. I can't speak about them much (defense projects, largely classified) but I can say that one involves natural language processing (on small datasets , small by internet scale that is), the other involves robotics and computer vision (involves some fairly large datasets but I don't need specialized data store infrastructure).
The interesting thing about such projects is that they combine cutting edge technology with interesting users - soldiers, intelligence analysts etc - with interesting hard problems to solve. It is a different world from fixing bugs in the leasing system from hell (old time ThoughtWorkers will understand what I mean ;-) ).
I'm pretty excited by the fact that in the DB scene there are also new players in the closed source side. If we want to make sure that in the end we'll have the best solutions I guess different economic and development models should compete as well, not only technology. For instance if I remember correctly MySQL initially started as a closed source product.
You know C/C++ like the back of your hand.
So are they looking for C OR C++ programmers, either may apply? or are they looking for strong understanding of the nuances of both C and C++?
I encounter a lot of "C/C++" on resumes and after interviewing a lot of folks, I've grown a distaste for this joined tag. The problem is that when I see "C/C++", I never know until I interview a candidate whether the intent is that "C/C++" is...
1) Shorthand for I program C and I program C++ (I know both well)
2) I program C with classes. I don't really know OOP, but I fake it with big class buckets.
3) I program in C++ and know OOP well, but I would get lost in the strict procedural world of C.
If you don't get where I'm coming from, I've programmed in both for many years, and I will simply say that "C and C++ are two paradigms separated by a common language."