[PhillipT]

Usually asset class would be the investment strategy area the bank follows and would include stuff like equities, fixed income, convertible bonds, alternative investment, commodities etc (depending on the bank they may have more or different asset management classes). Most of these have dedicated systems and tool sets (meaning you could end up maintaining windows apps or having to learn Q or K languages - depending on bank and system you work on). If you are leaning to quant/math then commodities is a good area which will draw on quant/math background as opposed to back office which is more standard IT. Hope this helps.

[anonplease]

Thanks, that does help a lot.

Is it possible for me to guess what technologies are used in each asset class before I join? For example, if I propose to join commodities as you recommend it, is it typical for an investment bank to only use X, Y, Z technologies for this asset class?

My fear is being stuck in back office "IT". However, without actual working knowledge of an investment bank, I can't really say where is best to place me. It may seem trivial to want to avoid C#/.net and the likes, however probably as most hackers on here I'd prefer to work in a team that uses technologies that I'm accustomed to.

Reading "On becoming a Quant" by Mark Joshi (http://www.markjoshi.com/downloads/advice.pdf) has really helped. C++ seems to be an absolutely necessary skill.

[PhillipT]

Rather than guess - I'd recommend asking. Most places will be happy to explain what languages/tech they use to an incoming employee and many will be happy you asked. Ultimately they are recruiting you because of your skill profile and trying to find the best place for you since in the long run it will benefit them (and hopefully you). So ask your recruiting contact for more info - and when you do so mention what you have found in your research and readings - because it (a) shows your committed and (b) will help them quickly see what are relevant possibilities.

The quants i know in finance have varied backgrounds but are predominantly maths, comp.sci and physics backgrounds and all asked similar questions on entry into their banks and financial institutions. As for languages they work with a range of tools including Matlab ,Mathematica and K for those on research side to C++, java, Python (wrapping C++ libraries), K /Q and R for those on production and reporting side. (edit - there are some who also use .net - but i forgot about them :-)) What tech is used will really depend on the bank. As a quant your job will be to work out how to implement and maintain algorithm, systems and tools around their tech environment - so it is probably better to be open on the language side and look at it all as huge good learning opportunity (which it is) and find a spot where the subject matter is going to intrigue/interest you. Again hope this helps.

[gtani]

Q or K

reminds of my second dev job, mortgage and asset backeds at Merrill in APL2 on a huge IBM mainframe cluster. Sort of fun and painful at the same time, esp. since APL was a language I couldn't really explain in code review (kinda like haskell). Also it was 14 hours/day, 6-7 days /week, eat at your desk, take a limo home.

They made it easy to move around between groups and divisions but you had to pitch your new manager for a slot, you couldn't jsut say you wanted to work there, it was really enterpreneurial, and still is. Very demanding, good dev skills plus either good at linear algebra and Diff eq, or at reading thick legal docs and translating pretty convoluted asset pools and deal structures to the dollar.