Hacker News Reader

Comment by adrianN | Hacker News Reader

Maybe it depends on the type of software development that you want to do? What area is your expertise in?

I think I know quite a lot about C and C++ (which means Modern C++ (currently C++17)), but have also worked in a lot of other programming languages (including Python, Java, JavaScript, Erlang (additionally, some further ones for university courses)). Additionally to computer science, I studied mathematics, where I also have a doctorate. I wrote my doctoral thesis about some very tricky questions about geometric structures behind optimization algorithms for discrete optimization problems. In all of these areas (mathematics, computer science, doctorate in mathematics), I either graduated with best possible or second best possible mark.

I also claim to know more about IT security than my CV suggests (I got one job interview for a job in this sector, and afterwards the interviewer told me that I have far too much knowledge in this area to be suitable for the available jobs; "you belong into academia").

The same holds, I believe, for embedded programming, where I also think to know quite a lot. Why am I so sure about it? Because as a hobby, I worked on reverse-engineering the firmware of some quite non-trivial embedded device.

I also think that I have an immense knowledge of arcane details of the instruction encodings of various CPU architectures, but this is merely a hobby of mine and useless for the job market.

There exist lots of people who seriously could not understand how I could not find a job as a programmer. For example quite some doctoral classmates (who could find a job in software development) called me a "wandering encyclopedia of the history of computer science".

As I said: I found a job in a sector that is very different from software development. Just don't tell me that it is easy to find a job in software development; it isn't, even if you are willing to relocate anywhere in Germany or Austria. And I know quite some people who have the same experience.

If you're still looking for a C++ job in Berlin, drop me an email.

If you are talking about think-cell: about 1.5 years ago, I applied, but was rejected after the programming test (even though they liked my CV and had to admit that the C++ code that I sent in was of decent quality).

Let me put it this way: I am more of a deep thinker than a fast thinker and (to put it in mathematical terms) more of a theory maker than a problem solver (i.e. I prefer Alexander Grothendieck over Paul Erdős, even though, of course, both were great mathematicians).

So, I was not capable to solve the hard algorithmic problem that think-cell had in their programming test in the given 9 hours; not for a lack of C++ skills, but for a lack of experience in developing "crazy" data structures and rather being a deep thinker/theory maker instead of a fast thinker/problem solver.

Solving interview problems is something that can be trained fairly effectively. Investing a few dozen hours into doing that improves your chances in coding interviews a lot. I tanked my first job interview (at Google) because I didn't know how to prepare. The next interviews I did much better.

Don't take it the wrong way and don't take it personally, but herein may lie the problem: companies are rarely looking for a deep thinker, but for said problem solver and someone getting shit done. At least, that's what most software development is about. Building things that can be sold.

That is why I wrote "(to put it in mathematical terms)": also Grothendieck and his scholars solved hard problems.

If you compare mathematical style of the mentioned Alexander Grothendieck and Paul Erdős, you will sense that they represent very different styles of doing mathematics. That is what I am referring to.

Oh hey, I figured out why you can't get a job as a software developer. You didn't mention in your first post you refuse to get good at interviews.

Solving hard algorithmic problems in a job interview is a very uncommon interview practice in Germany. In all other job interviews that I had, the programming questions were ridiculously easy. UPDATE: I actually got rejected in one job interview because I knew too much about some quite subtle details of the C++ language (claimed reason: I am not "sufficiently pragmatic" as a programmer).

But I actually did solve some really hard algorithmic questions in my doctoral thesis - they are just of a very different kind than the "algorithmic brainteasers" that seem to be common in job interviews in the Silicon Valley.

I'd say for most jobs you are overqualified and people are afraid you would be bored after some time. If I were you I'd apply at an institute like the DFKI, I'm sure they would be glad to have you and provide you with a good challenge.

The things recruiters told you ("not sufficiently pragmatic" or that you belong in academia)... they didn't tell you this to talk down to you but to give you hints.

> The things recruiters told you ("not sufficiently pragmatic" or that you belong in academia)... they didn't tell you this to talk down to you but to give you hints.

I would have loved to stay in academia. The problem, why I left academia is the infamous Wissenschaftszeitvertragsgesetz, which gives you an employment ban on temporary contracts in academia after 6+6 years. Since nearly every employment contract in science is temporary, this actually implies a nearly complete employment ban in science after at most 12 years.

I think you can stay longer if your position is paid for by a private party instead of publicly funded. That could very well be the case at an institute. Maybe reach out to them, best a professor and explain your situation.

Link in german:

https://www.forschung-und-lehre.de/die-zehn-haeufigsten-irrt...

Nr 3

If you are around Munich, still looking for a job, drop me a line at info at langsys.com

Deep thinking is what we look for.

While I am not currently looking for a job anymore (as I wrote: I found a job in a completely different sector), I nevertheless tried to find out what the company langsys is doing (since it is not unccommon for me to meet other people who have/had very similar problems). Unluckily, the website https://langsys.com/ contains no information.

Reply on news.ycombinator.com