Which technology-related companies do the most interesting work?

Which technology-related companies do the most interesting work?
What skills are the most in-demand at those companies? Cybersec? Statistics?

Other urls found in this thread:

stackoverflow.com/jobs/142391/sr-software-engineer-core-platforms-jp-morgan-chase?so=i&sec=False&pg=1&offset=-1
glassdoor.com/Interview/BlackRock-Interview-Questions-E9331.htm
glassdoor.com/Interview/Renaissance-Technologies-LLC-Interview-Questions-E19369.htm
wallstreetoasis.com/forums/getting-into-rentech
norvig.com/21-days.html
twitter.com/NSFWRedditVideo

Amazon because of AWS. All sorts of skills are in demand I guess

FinTech is an emerging market, so I'd be interested in that. There are newly minted positions like "Quantitative Developer" that are not just catch phrases, but actually very specific job titles.

Not just any software engineer can grab those jobs. You need to be able to code and implement stochastic models, math concepts, etc. I'd look into that stuff.

Those jobs require a very high level in math, right? Isn't it more about mathematics than programming at that point?

That's the tricky thing about it.

It's high level math, but it isn't more about mathematics than programming. You need to know high level software engineering to also implement the high level mathematics. The person presumably hired for that position needs to pretty much be a fucking expert at both fields.

For example: If I ask an elite software engineer to implement the Black Scholes model, he won't know what the fuck to do or how to implement it because he doesn't understand the mathematics.

On the other hand, if I ask a mathematician to do it, he won't know how to code it, despite knowing how the black scholes model works.

As you can see, it's a catch-22. It's an emerging market filled up by people with intensive math+compsci backgrounds or financial engineering backgrounds.

It's useful to be aware that a typical financial engineering curriculum is just high-end math with coding and financial terms thrown in.

>Palantir

Pretty cool software. I'm surprised that some people on Sup Forums didn't create a clone already

>a fucking expert at both fields
Unfortunately I have absolutely no natural aptitude at math and I've stopped studying it after graduating high school. So, should I forget about financial tech?

How is it able to track bombs and terrorists on battlegrounds?
>big data is a meme they said

>For example: If I ask an elite software engineer to implement the Black Scholes model, he won't know what the fuck to do or how to implement it because he doesn't understand the mathematics.
If you ask even a good software engineer (not even saying elite), they will be able to learn. CS degrees typically include a semester or so of statistics. A good university education teaches you how to research and teach yourself.

Yes, there is some domain specific knowledge, but that's the case in all private sector programming jobs.

With all that said, I don't see anything there that's beyond what a semester of stats at a good school covers.

As for the PDEs, a good CS education includes a semester on differential equations, which includes PDEs. You would typically not write your own PDE solver anyway.

>Black Scholes
>Semester of stats.

It's typically graduate level coursework. It's rare that undergrads will be introduced to it unless they're on a financial engineering track. It also requires in depth knowledge of stochastic PDEs, background knowledge of Brownian motion, etc.

>A good university education teaches you how to research and teach yourself.
But yes, I generally agree with this.

The PDEs learned in undergrad are trivial compared to graduate level coursework.

Of course not! The FinTech sector is brimming with tons of possibilities for several software engineers, it's just that the QuantDev jobs are slowly becoming the most in demand.

In general, I think FinTech is going to take off. If/when you get a chance, check out Robin Hood and Stash. They're two huge players in the FinTech scene right now. People are starting to get interested.

Send a resume over to a FinTech startup or firm. Fucks sake, this is an ad for JP Morgan asking for a Software Engineer:
stackoverflow.com/jobs/142391/sr-software-engineer-core-platforms-jp-morgan-chase?so=i&sec=False&pg=1&offset=-1

>We are looking for:

>Hands-on, technology-focused developers, who enjoys working through hard technical problems.
>10-15+ years of hands-on Application Development work experience.
>Smart developers with track record of delivery, whether in previous professional roles, college or personal projects (for instance open source software contributions.
>Strong development skills across platforms and languages.
>Experience with popular programming languages such as C++, Java, C# and/or Python is beneficial, but not required.
>A strong comprehension of Computer Science fundamentals (data structures, algorithms, graph traversal, design patterns, interface design) is preferred over specific experience with a particular programming language.

>QuantDev jobs are slowly becoming the most in demand
They're not though. It's a very small job market with ruthless competition. What's currently happening is QA becomes a meme buzzword like Data Science where every two bit statistician is a ""data scientist"".

>data structures, algorithms, graph traversal, design patterns, interface design
Not the same guy, what are some good resources to get a solid understanding of those? I know how to program, but my code is probably shitty due to my lack of formal CS education.

Good point. The authentic QA positions are becoming most in demand in the FinTech market, though; you wouldn't agree with that? Just in that slice of the market, not the whole pie.

I personally know a lot of friends who tried to get into BlackRock, but couldn't get past some of the technical questions they ask. They apparently start off with technical, applied comp sci stuff, then start asking about theoretical math/comp sci.

I liked The Algorithm Design Manual a lot, it's not too dense and gives a good overview. Don't forget to do the exercises

Thanks.

Troll around your public university's CS website, troll professor pages, look through their 'courses' section and research all of the books and complete their assignments. I was in a similar position and that's exactly what I did.

Some professors are baller as fuck and put up their entire lecture notes, plus homework+solutions, PLUS exams/quizzes+solutions, along with their syllabus. Some even point you to alternative sources of education, which is even better. It's getting a 10k+/yr education for free. It's incredible.

>BlackRock
What kind of people even manage to make it through the interview process with these companies? I heard Rentech required PhDs for internships.

Autists or very, very hardcore engineers.

The two people I know who got into BlackRock were BS/MS students, with one considering going to graduate school for his PhD. Both were STEM majors; Physics and EE, respectively. Both were insanely smart and talented, but all they did was study. Saw them in the morning, afternoon and evening in the library.

It wasn't so much that they had so much innate talent, but rather that they took their classes/future careers extremely seriously and treated them as jobs. They would come to the library around 8-9am, study, go to class, eat, study, go to class, study, eat and sleep.

Pay is fucking insane, too.

>authentic QA positions are becoming most in demand in the FinTech market
I wouldn't say so. Derivatives pricing and asset management is a small part of overall fintech sphere and while QA techniques have many more applications they rarely are as effective and have lower RoI elsewhere. They're "in demand" in the sense that there's a shortage of specialists, mainly because it's not the most exciting job for high-level mathematicians or computer scientists and not the easiest one to get into for "pure" developers. It certainly doesn't help that it's a front office workload/hours couple with back office pay. There's more than a few former quants in management consulting/non-tech finance where you still work like mule, but at least you get an appropriate compensation and career prospects.

>tfw doomed to mediocrity because not autistic or hardcore enough

>It's typically graduate level coursework. It's rare that undergrads will be introduced to it unless they're on a financial engineering track. It also requires in depth knowledge of stochastic PDEs, background knowledge of Brownian motion, etc.
I realize that the model itself isn't covered in first semester undergrad stats, but all the things used in it (PDEs, distribution functions, stochastic processes and modelling them) are all covered.

>The PDEs learned in undergrad are trivial compared to graduate level coursework.
I understand that, but there's nothing particularly advanced in those PDEs. You'd typically use a library to solve the PDEs numerically, anyway.

The best way would be to get a CS degree from a good school. The second best way would be to look at the syllabus for a top school and learn all of their material and do all of their assignments. If you have access to a good CC, take their algorithms/data structures course, or sit in on your local university's course.

Becoming hardcore is a matter of mindset and will, not autism.

If you dedicate 7+ hours a day to coding for about 2 months, I can almost guarantee that you'll get an entry level position and work from there.
Damn, you seem to know a lot about the sector. Are you a quant or trader?

glassdoor.com/Interview/BlackRock-Interview-Questions-E9331.htm

Read the questions. How hashmaps avoid collisions (specific techniques), how java strings work, interfaces vs abstract classes are all things that I (really, really) hope a BS CS or talented AS CS graduate could tell me.

glassdoor.com/Interview/Renaissance-Technologies-LLC-Interview-Questions-E19369.htm
Look at those questions. Does that seem impossible?

>If you dedicate 7+ hours a day to coding for about 2 months
I can do that. I almost had that rythm when I was learning C (I worked all day long for around a month). But that'll just make me a better programmer, is it enough?

I'm in math research, but have prior experience and many friends in finance and fintech specifically.

>Difficulty 3.3
>for fucking Rentech
What?
wallstreetoasis.com/forums/getting-into-rentech (I know, >WSO, but still)

norvig.com/21-days.html

>wallstreetoasis.com/forums/getting-into-rentech (I know, >WSO, but still)
Ah, I was just talking about the questions. Getting an interview at any top shop is a whole different ball game. The best technique is always to know someone.

I did a phone interview for a SRE position here a while back (did horribly)

most of the questions were about java / tomcat and troubleshooting that shit which I've never really done
seemed odd, but still a company I would have worked for

>I know, >WSO, but still
Not still. Most business/finance majors are dumb as fuck when it comes to science, so it's no surprise that Calc III material is seen as bleeding edge triple-PhD knowledge.

>Getting an interview at any top shop is a whole different ball game.
Is the process different between hedge funds and smaller HFT/algo shops?
It's still observable that such firms hire PhDs almost always, if not exclusively.

>such firms hire PhDs almost always
Indeed. Those are the people rating the interview questions on glassdoor.

Looking at those interview questions, it's really not that different from what you'd encounter at interviews for "elite" tech shops (amazon, google, etc). Which questions are giving you trouble?

What kind of CS PhDs are they looking for? I'm assuming machine learning.

Specific field of research is rarely relevant outside of actual research positions.

What are you implying? It takes time to get good in anything, that's a given.

Is there a place in fintech for cybersecurity professionals?

doesn't palantir only have like 100 employees? i'd imagine they'd all be PHD's or former google employees with 10 years experience

>Stochastic differential equation modelling and Ito's Calculus with only one semester of stats