• C++ Programming for Financial Engineering
    Highly recommended by thousands of MFE students. Covers essential C++ topics with applications to financial engineering. Learn more Join!
    Python for Finance with Intro to Data Science
    Gain practical understanding of Python to read, understand, and write professional Python code for your first day on the job. Learn more Join!
    An Intuition-Based Options Primer for FE
    Ideal for entry level positions interviews and graduate studies, specializing in options trading arbitrage and options valuation models. Learn more Join!

Type of internship for wannabe quant

Joined
12/28/10
Messages
84
Points
18
Myself: Junior at Johns Hopkins, Applied Math + Economics (gonna start taking comp sci courses this fall), 2 internships in finance (non-BB banks) so far (doing one right now).

I have a 3.7 at a school that doesn't grade inflate to the degree some comparable schools do (Brown, Yale avg gpa 3.6, Hopkins 3.18), and I feel like human resources at some firms don't realize how important this is, but I digress (hugest pet peeve of mine).

I was wondering what types of internships a wannabe quant (probably not interested in model validation, algorithmic agency trading, more interested in quant. investment mgmt, development) should pursue....
 
First, I don't believe you know what type of internship you will like or not until you get there. Some places the role will be well-defined, others you will be jack of all trades. In most case, you have to be at least usable with coding. If you do none of the coding, prototyping, then the role shouldn't be called anything "quant" at all.
Development: companies like Numerix/FINCAD make analytic tools used by quantitative side of the banks. Almost every financial firms have development groups that can use more good coder. It's a good place to hone your ninja skills.
Here is the list of companies with quant internship http://www.quantnet.com/quant-internships-graduate-recruitment-list-firms/
 
As a follow up, do you think a well established S&T internship at a bulge bracket would be ideal, or a more programming heavy role at a development shop?
 
As a follow up, do you think a well established S&T internship at a bulge bracket would be ideal, or a more programming heavy role at a development shop?

Well, seeing as how sales and trading isn't quant, a programming heavy role would be ideal if you want to some day be knighted in quanthood.
 
Quantitative investment management and development are two very different things. Which do you want?
 
I don't really know. I want to keep options open, but I think I would be more interested in quant. investment mgmt.

I don't see why you immediately discredit S&T; I see it as a good introduction to the markets. You can largely learn programming on your own, however, getting a real hands on experience in the markets is not as easy to acquire.
 
I don't really know. I want to keep options open, but I think I would be more interested in quant. investment mgmt.

I don't see why you immediately discredit S&T; I see it as a good introduction to the markets. You can largely learn programming on your own, however, getting a real hands on experience in the markets is not as easy to acquire.

You asked which is ideal for becoming a quant later and I gave the answer. Getting hands-on experience programming with financial data isn't exactly easy to acquire either. This is the most important skill for a quant.
 
You asked which is ideal for becoming a quant later and I gave the answer. Getting hands-on experience programming with financial data isn't exactly easy to acquire either. This is the most important skill for a quant.
This can vary depending on what flavor of quant you want to become. However, it appears that the derivatives heydey is gone... so the jobs will be either high frequency trading or risk management.

In the HFT space, you are dealing with massive amounts of data... on the order of 10+GB/day. Even being able to load this data into a program and clean it can be a challenge, not to mention the analysis you have to do.
 
This can vary depending on what flavor of quant you want to become. However, it appears that the derivatives heydey is gone... so the jobs will be either high frequency trading or risk management.

In the HFT space, you are dealing with massive amounts of data... on the order of 10+GB/day. Even being able to load this data into a program and clean it can be a challenge, not to mention the analysis you have to do.

I'm interning at a start up where I'm experiencing the "joys" of extracting and cleaning data. Sometimes cleaning the data is half the work. I'm dealing with mostly unstructured data (text). I apply machine learning algos on the data
 
This can vary depending on what flavor of quant you want to become. However, it appears that the derivatives heydey is gone... so the jobs will be either high frequency trading or risk management.

In the HFT space, you are dealing with massive amounts of data... on the order of 10+GB/day. Even being able to load this data into a program and clean it can be a challenge, not to mention the analysis you have to do.

Yeah, what I said was conditioned on derivatives creation being pretty much dead at this point. This is why the 90s were dominated by physicists and the 10s will be dominated by engineers and computer scientists.
 
I'm interning at a start up where I'm experiencing the "joys" of extracting and cleaning data. Sometimes cleaning the data is half the work. I'm dealing with mostly unstructured data (text). I apply machine learning algos on the data
Wow, you must be dealing with a ton of data. In my trading research internship 80% of it is coding to get the data ready, 10% of it is coding up the actual algorithms, and 10% is "research". The importance of being able to code efficiently and deal with massive amounts of data cannot be understated...
 
Wow, you must be dealing with a ton of data. In my trading research internship 80% of it is coding to get the data ready, 10% of it is coding up the actual algorithms, and 10% is "research". The importance of being able to code efficiently and deal with massive amounts of data cannot be understated...


I'm dealing with relatively sized data, but it's mostly text, and I'm having to clean / transform / parse and vectorize text .

My entire db corpus has north of 35million tokens. I think this is a couple of GBs and counting.

I'm curious what tools you use to get the job done. I'm having to do quite a bit of natrural language processing, regular expressions, implement machine learning algos, computations with huge sparse matrices ( think dimensions - 2000 x 32000 ), etc , so I'm using Python as my primary language. I find myself going to R, SQL, JMP and a little SAS as needed.
 
Back
Top