Forums  > Careers  > switch from quant to machine learning/IT  
     
Page 1 of 2Goto to page: [1], 2 Next
Display using:  

prokyons


Total Posts: 17
Joined: Oct 2012
 
Posted: 2012-10-24 04:29
The following post isn't meant to offend anyone who is currently working as a quant, I just want some advice regarding my current situation.

About me: MSc from a top school in computer science and a pretty solid statistics and programming background, plus 1.5 years working as a quant in market risk for a large bank. I like my colleagues, the pay is great, and the work is relatively interesting: I get to build and implement new models (although basic in nature, I still find them interesting), I work with traders, etc.

I also have a fairly good grasp of machine learning algorithms, and made it through 3 rounds of interviews at top IT firms. So, from a cynical perspective, if I want to quit finance then I sure can.

The problem: I just don't see how what I do in finance on a daily basis benefits anyone. The actual models which we come up with are no-where close in estimating the real risk (in my opinion). It's almost as if we're playing charades, and are creating models as an elaborate smoke-screen for the regulators. We don't create stress tests. We don't think of creative market scenarios where shit can really hit the fan. We don't work with traders to the level of influencing their trading decision. The list goes on.

I've thought of switching fields within finance, to even something like pricing. I hear similar negative "why are we doing this" messages from other friends who are in pricing; most common seems to be "why risk neutral and not risk averse pricing models?". Those who work with traders claim that they're only doing this for the money because traders have no souls (they are not fully enjoying their work days).

Perhaps I am being too cynical, but in summary, most people whom I know are really just being quants because of the money: they needed a job, banks had more openings than their industries and they made the switch.

In my case, however, I could either go as a programmer for a top IT firm, or as a machine learning scientist. The pay which I got offered is about the same, and what I like about the job is that they are actually building a working system, which does stuff. They are manufacturing something, not creating a mathematical smokescreen for the financial regulators. They deliver a working product.

So, should I try working in a different area? I figure I can always come back to be a quant.

Thanks for all your input!

goldorak


Total Posts: 442
Joined: Nov 2004
 
Posted: 2012-10-24 05:20
go for it.

If you are not living on the edge you are taking up too much space.

dgn2


Total Posts: 1924
Joined: May 2004
 
Posted: 2012-10-24 05:26
I say go for it as well. Sounds like you are pretty early in your career so you probably have little to lose.

...WARNING: I am an optimal f'er

andre_the_giant


Total Posts: 14
Joined: Apr 2010
 
Posted: 2012-10-24 09:34
I would say that IT isn't your only option. You can also go for hybrid projects, like a startup that is looking for quant/risk modeling/developers for a disruptive product. Lots of this kind of thing is going on in California right now.

As an example, look at the quantitative modeler:
https://expedite.cc/jobs

I know two of these guys behind this - both very smart and early paypal engineers. My only connection is that one of them is an investor in my company.

sv507


Total Posts: 138
Joined: Aug 2010
 
Posted: 2012-10-24 11:15
prokyons

I sympathise with what you are saying. However, I think you are not getting the whole picture. Somewhere in your bank there will certainly be stress tests being made. The traders and risk managers job is to identify the stress scenarios and estimate their losses - and its they who add in the risk aversion. Sadly most quants do not want to get involved with the reality of the market.

My impression of the ML job market was that you are just sucking out click data to predict what someone will buy. I'm not convinced that it really works, and its hard to argue that its benefitting anyone either. [capafan on W****tt is doing this work...maybe you want to contact him]

I guess the thing I would be more concerned about is that the derivatives market is shrinking with increased capital requirements, so jobs/salaries will decline in the coming years.

gentinex


Total Posts: 60
Joined: May 2006
 
Posted: 2012-10-24 14:28
Like sv507, I'd say to be careful about 'grass is greener' thoughts..I do question whether a lot of the work being done in the 'big data' space is legitimately benefiting society, and it's not necessarily easy to find the kind of work that really does.

I do agree that what you're doing now is likely just about gaming regulators. But if you're only talking to colleagues and pricing quants at large banks, you are seeing a very limited picture of what's out there in finance. There are more interesting things, but as with tech or any other industry, it takes some effort to find them - even more so now (especially compared to tech, as you seem to be discovering) because the stricter regulatory environment means that all large banks are hiring more quantitative people for regulatory risk work than for other kinds of things.

granchio


Total Posts: 1466
Joined: Apr 2004
 
Posted: 2012-10-24 15:17
I agree with others that there are places in finance where people do so-called "useful" work, not smokescreens.

But I still think you should switch: there are too many people in quantitative finance and not enough "useful" jobs. That is the advice I give to most young/juniors who want to get into quant fin now: don't - unless you really, really love it.

"Deserve got nothing to do with it" - Clint

prokyons


Total Posts: 17
Joined: Oct 2012
 
Posted: 2012-10-24 16:51
@dgn2: My thoughts exactly. I won't know for sure until I try.

prokyons


Total Posts: 17
Joined: Oct 2012
 
Posted: 2012-10-24 16:57
@andre_the_giant: Buddy, I'm very interested! You've hit the sweet spot. I'd love to relocate to California too. Time to start brushing up on my comp sci / ML background :p

Another application I've been thinking about is machine vision to unmanned aerial vehicles. The number of applications is virtually unlimited: surveillance, military, flight path correction in commercial aircraft, etc... OK, so I have to flight-related hobbies on the side too.

prokyons


Total Posts: 17
Joined: Oct 2012
 
Posted: 2012-10-24 17:04
@sv507: I've heard of pensions funds making really cool econometrics models, but they are looking for someone with MSc/PhD in finance usually, or in economics. The job is interesting from that perspective, but is lacking on the coding / math. But it's certainly a lot more interesting, probably #1 on my list of finance jobs.

In most banks that I know of, stress tests exist for a check mark: no-one really looks at them, and traders have their own. Where I work, half the stress tests are broken, and no-one cares because their output is never really reported anyway.

Thanks for the contact, I'll ask capafan.

Yep, our desks are being pressured like crazy to increase capital. I agree that derivatives market is shrinking.

prokyons


Total Posts: 17
Joined: Oct 2012
 
Posted: 2012-10-24 17:21
@gentinex: Yes, e-commerce can be compared to algo-trading: it's luring consumer fish to spend money on some crap they don't really need. Kindle is a good case study :)

Could you give me some examples of more interesting things in finance?

I'm thinking of this as a stepping stone. I get to improve my programming background, my ML algo background and I save more money and get a loud company name on my resume. If I want to apply my skills to curing cancer or creating some piece of technology which benefits people in any way, then I am a lot closer to doing it than coming straight out of finance.

Apart from consulting (which is not really scalable and is tiring), what kind of a business can one create with 10 years of work experience at a bank?


prokyons


Total Posts: 17
Joined: Oct 2012
 
Posted: 2012-10-24 17:29
@granchio: yep, agree 100%. Same question as for gentinex: which areas in finance are more interesting?

For the record, I know people with BEng, MBA and CFA, who are one position higher than me in their group who are being paid the same as me! A secretary dumped a file with everyone's salaries under her VP in a public directory by mistake. Lot of people were pissed :) So yeah, finance these days is over-saturated with degree inflation.

IVolrev


Total Posts: 76
Joined: Oct 2007
 
Posted: 2012-10-25 04:13
My advice: Do what you think you love and have a strong passion for. If it is to create something, to be innovative with the purpose to benefit others (innovating CDO cubed does not count as such pursuit) then you are not belonging into a financial firm.

From your explanations it sounds that despite your "top school" you did not really end up in the "top spot" of your bank. At the same time you do not seem to be too hungry nor passionate about the job as trader. Thus, I would reckon that you will not be satisfied with other roles in the bank because you described what you are enthusiastic about and aside the job as trader I am not sure which part of an investment bank one can be totally ecstatic about (and nowadays being a trader in a bank is really quite a boring pursuit and often waste of time for sake of very expensive opportunity costs).

Thus, I believe you want to look for a job that fills you out more than sitting at a desk without much of future opportunities to step up to something more challenging and important (I know what you are talking about in terms of smoke screens and I just do not see where this may lead to later on). Sorry for being so blunt, I see things from a trading perspective and I would do this job even with zero income and you talked a little too much about money, income, pay which tells me material things matter to you more than the pure passion for what you do in life. I believe money takes care of itself once you find a job that fulfills you and that you are good at as a result of your enthusiasm and love for the job. Good luck!!!

prokyons


Total Posts: 17
Joined: Oct 2012
 
Posted: 2012-10-25 17:06
@IVolrev: interesting you should bring up material things. I have a lot of interests, which I split into two categories: those which bring money and those which don't. The ones in the latter category are hobbies, and the ones in the former I look for jobs with. I'm not saying this is the right thing to do, this is simply what I am doing (feel free to disagree).

To give a concrete example, I like compute programming/algorithms/machine learning (basically most things which are computer-science related), fitness training and I also have an obsession with RC aircraft.

Most high-paying jobs are in computer science, albeit the cons which come with the job: sitting all day staring at a computer screen (bad for lower back and eye-sight), not getting out to a lot of meetings (low social factor) and working in a male-dominated field (before you're married, this can also be seen as a drawback, because you simply do not meet a lot of women on a daily basis; none actually).

Now consider my other hobbies. I've worked as a personal trainer in my first two years of undergrad. Those jobs are hard to keep, the pay is low and unless you have the genetics to look like out of this world (yes, what you look like is a huge factor in the fitness industry), it's also harder to get new clients. On the upside, you meet a lot of good-looking women, the social factor is amazing and the job is relatively healthy. Big trade-off.

I'm thinking of getting back to personal training on weekends, as a second job, but mostly for the fun and the social factor... I'd be making $200/weekend if I'm lucky, after the gym takes their cut.

The RC aircraft hobby drains more money than I can get out of it. Unless I become a supplier/distributor and open my own hobby store (which is really far-fetched), I don't see any way of making money from this. And if I really want to become a distributor, then I'm much better off selling something which people crave in larger quantities, like water for construction sites, baby foods, etc... now we're on a tangent of opening my own supply chain business, which is a whole different topic.

mabcta


Total Posts: 4
Joined: Oct 2012
 
Posted: 2012-10-25 18:01

Awesome discussion and great insights from great people. Couldn't resist jumping in.

Helping general public manage their financal futures (genuinely try and succeed to make them some money) is a worthy cause, indeed. But, yes...that's not what you do at most quant shops. May be you could become an RIA/CTA and manage individual's accounts. Not easy at the outset, but very feasible - you don't need $100M to start a hedge fund to help community with your finance/sharp IQs and other skills set.

For all those quants that have strong technical/modeling skills - how cofident are you in your "predictive modeling" abilities? NOT building mindless, bruteforce machines that make money off of nano-second advantage (something that will and should eventually become illegal), but some predictive power and with taking some true financial risk in anticipation of some reward. If you are, AND if you are not satisfied with your typical quant "job"...why don't you join/start a firm that actually manages/advises individual investors' money? IF Hedge funds are not soul-satisfying to you, but investment management/financial markets is, then what are you thinking about doing? Can you put your money/career where your mouth/mind/heart is?

That's what the OP needs to answer to himself. If you want to do something more meaningful, how about wilingness to try to apply your skills to building models to find potential accounting issues from 10-Qs (forensic accounting)? If this sounds interesting, I can put you in touch with someone with a startup business model (but, not much startup capital yet) and you can explore if you can be a part of some social good that also could make you a lot of dough IF succeeds...and, may be have a lot of fun/challenge doing it.

Why do we all think that someone else should have a working company, willing to pay us a decent amount for having fun and in our search for our self actualization? If you have earned decently and have saved wisely, you should have enough risk capital to go out and try to do something for this world on your own...or, look to join forces with someone who might be on that path - taking some risk. The reward of "satisfaction" (at any level) is much higher, so it should follow that we can't expect it coming from a relatively low risk "job/career". What are you willing to give up and give in to explore your heart's path?


IVolrev


Total Posts: 76
Joined: Oct 2007
 
Posted: 2012-10-26 03:32
@Prokyons, your response to my post only shows clearly that you will not be happy within financial services. So, yes I agree with some previous posters. Go for it, make a change in your life. You seem extremely picky and sometimes contradictory: You say on one hand a big down within CS jobs is that they are male oriented and its hard to meet women. However, why do you bother if you are able to meet them elsewhere. You brought up the fitness trainer thing and even elaborated how someone needs to have the looks to attract clients. So, where is the problem here? I am getting kinda confused what you actually want...also it sounds contradictory that on one hand you compare salaries in competitive IT jobs vs salaries on the quant side. Then suddenly you start talking about $200/weekend kind of jobs. Pretty unfocused if you ask me. Maybe you wanna think what really matters to you in life before you think about which job will allow you come as close as reasonably possible to the actualization of those wants.

I can only repeat myself: Go for what you have a passion for, the money, income, women, all else will take care of itself.

jslade


Total Posts: 831
Joined: Feb 2007
 
Posted: 2012-10-27 01:40
"To give a concrete example, I like compute programming/algorithms/machine learning (basically most things which are computer-science related), fitness training and I also have an obsession with RC aircraft. "

The CIA/MI6 is hiring. It was not for me, but it's the only thing I can think of which combines those interests.

I consult as a ML/Forecasting guy for various startups and financially places. It's aite. Probably more fun than coming up with regulatory risk models. On the other hand, most work is meaningless. If you stick with a company in one role for long enough, you will get bored and jaded (one of the reasons I consult), no matter what they're selling.

Drop me a CV if you want a California relo. Most of my customers are hiring.

"Learning, n. The kind of ignorance distinguishing the studious."

prokyons


Total Posts: 17
Joined: Oct 2012
 
Posted: 2012-10-27 18:36
@mabcta: I'm not much of an accountant :) I'm sure it's a fine profession, and managing people's financial futures for the better is a rewarding experience, but it's not for me.

You mean private equity? I've spoken to a few folks who are in the biz and to some who are trying to get there. First, it takes time to build a client base. Second, most people there are MBAs/CFAs. I think that some material in CFA is interesting, but most of it I find really boring. It's not very quantitative, and one can do all three levels just by brute memorization. That's not to say that there are people who deeply understand all of CFA material, whom I respect deeply. Anyway, I guess my point is that I don't find CFA/MBA/FRM/PRM material nearly as interesting as computer-science and applied math.

Forensic accounting sounds like an area where one can use a lot of machine learning, for example fraud detection. Then again, I bet 95% of frauds can be located with simple deterministic algorithms. I can't send you a PM, your e-mail does not show in your profile :) Can you PM me?

Good point about risk in career. I've been doing quite a lot of research regarding starting my own business lately, which only suggests that I should get more into IT field, because skills there are hugely beneficial in starting my own business.

prokyons


Total Posts: 17
Joined: Oct 2012
 
Posted: 2012-10-27 18:41
@IVolrev: good points. I only said $200/weekend to suggest that I can do this part-time, if I really like it so much (it'll always be an option). In other words, it's an example which suggests that I should not be doing it.

I enjoy coding design patterns/computer science algos/machine learning/applied math/stats. So, I guess that's the passion, and I'm better off moving to California (I love the weather there), because there are a lot of smaller companies there who need people with my skillset.

I've also realized that I need to brush up on J2EE and Hadoop if I want to make the switch. A lot of jobs are asking me for that skillset.

prokyons


Total Posts: 17
Joined: Oct 2012
 
Posted: 2012-10-27 18:47
@jslade: I'm Canadian, eh? :) Already checked my options with CSS a few years back... their "training program" did not appeal to me. Plus, for some reason they want PhDs. Doesn't matter in what, or from which school, just a PhD. One of their "top guys" (who claimed he was a top guy) had some English or history PhD from Trent (it's a small school in the countryside). So yeah... closed that chapter quickly. Never saw so many narrow-minded people in one spot in one time.

I'll send you a PM shortly, thanks for the hook-up!

astar


Total Posts: 173
Joined: Mar 2007
 
Posted: 2012-10-28 02:58
You have heard some very good advice here. I switched from a derivative / quant position to a tech startup / big data type role. Lots of machine learning, hadoop type distributed computing in some places. Cash pay is lower. Hopefully the equity will pay off. The kind of people are way different; in a good way.

Most tech firms, though, will probably look at someone moving from a finance background in a slightly different light. The coding styles and focus are different. Also, be careful of the "grass is greener" syndrome. Many "Data Science" roles could be pretty boring too (I see that sv500 mentioned that below). Many tech startups tend to view these groups as some exotic must-haves, without a very clear picture of what it is. NYC is probably a better place to run into some of these startups.

From your description it looks like that is not the case here, but you have to ensure that the business side knows something about what you will achieve; that there are clear achievable goals; and that your role is essential to the business.

Hope this helps, and feel free to contact off forum.

prokyons


Total Posts: 17
Joined: Oct 2012
 
Posted: 2012-10-28 17:37
@astar: yeah, I hear you on the people part.

How did you make the switch?

I feel like _if_ the business is making money on the IT component which you're involved in, then it's the equivalent to working for front office in finance: what you do can directly affect the revenue. Having said that, do data guys have a direct involvement into the money making part in companies like Amazon? I think a new business product like Kindle will bring Amazon far more income than the tweak on some algorithm which matches users and adds, no?

On the other hand, here's an example of a database company. They obviously are making a database. Which they sell, along with support, for money. So, if one is working on the database development then one is directly involved in the money-making part of the business. Much like the electrical engineers who've put Kindle together.

Then, are there data mining / machine learning jobs which do directly influence the business?

PM sent.

sv507


Total Posts: 138
Joined: Aug 2010
 
Posted: 2012-10-29 13:23
prokyons

I am no expert but look at eg amazon kindle cost
amazon is selling kindle fire below cost price with the aim of generating sales on (non) digital content.

Google is an advertising company


advertising really is the big driving force for these technology companies... but that money then allows google to research speech recognition/ driverless cars/ etc


sv507


Total Posts: 138
Joined: Aug 2010
 
Posted: 2012-10-29 13:57
@astar,jslade


Is there an equivalent of nuclearphynance for ML types?

astar


Total Posts: 173
Joined: Mar 2007
 
Posted: 2012-10-29 17:44
@sv no idea. perhaps some google groups / meetup discussions? I found the meetup discussions somewhat more suited for beginners, nothing similar to a pithy professional posts between practising experts. NP is probably the best place, and how many different phorums do you have time to read?
Previous Thread :: Next Thread 
Page 1 of 2Goto to page: [1], 2 Next