I am a Quant C++ Dev in NYC, working closely with a large quant team and loving it (mostly). AMA

Feel free to post questions below, and I'll try to address them as well as I can

 

[Daniel Duffy]

Can you please tell us some blogs or documentations or tutorials for experienced java developers to transition to Quant C++ developer ,apart from the Quatnet Courses?

My 2 cents.

A Java background is kind of useful for C++. But it is built on a completely different design. In a sense, it's like starting all over again.
TBH, I did a Java course in 1997 and I concluded (for several reasons) it was not for me. For maths, it is a nightmare. And the design of generics in Java is a bad fix.

I see Java programmers learning C++ using new left, right and centre but without delete.

I would say do QN C++.

 

[Daniel Duffy]

As a student, at the time I'd say:

Most
1. Fun: lambda programming and the final projects
2. Useful: templates and meta programming
3. Difficult: Design Patterns with threads

Least:
A. Fun: Type traits and decltype
B. Useful: Quizzes
C. Difficult: STL

some 1 line feedback from the viewpoint of C++20 and the advanced course.

1. Lambdas are cute in the beginning for initialisation of objects and thread start function in main(). For the rest, not so useful.
1. Projects are fun.
2. C++20 has a lot of nice templarte stuff (e.g. Concepts).
3. Probably the wrong paradigm. Micro-optimisation and flaky. C++11 has std::future and coarse parallelisation.
C++ multihreading and tasking is for experienced C++ experts.

A. C++11 was a new, major relese. Some of the features obey Pareto 20/80 rule. Life in C++20 is easier.
B. Quizzes can be useful when they force you to think "laterally".
C. STL is built on functional programming and has clean interfaces.

 

[Pavlos Sakoglou]

Thank you Pavlos for coming back and sharing your experience. We appreciate you, sir.
I'm featuring this on the homepage so this can get more attention.

Now, what would you do differently if you went back and were in the positions of many members here, planning to get a finance job? How would you prepare knowing what you know now.

Thanks Andy!

What I'd do differently:
1. Work on more personal projects in detail to compensate for luck of experience
2. Focus on mastering one concept at a time instead of being mediocre in many
3. Schedule fewer interviews and take my time to prepare for each
4. Seek advice and help from people in the industry I am preparing for

 

[Tigga]

Hey Pavlos,
What is your plan for the next few years? Is there any programming language, tech stack or any interesting trend you want to follow?

 

[Pavlos Sakoglou]

Hey Pavlos,
What is your plan for the next few years? Is there any programming language, tech stack or any interesting trend you want to follow?

I'd be interested in distributed computing, decentralized applications, and P2P architectures in general. I feel the world is moving towards that direction, especially large systems.

 

[MichaelLewis]

I am starting to wonder when I’ll need to learn FPGAs as well. Anyone else feel the same?

 

[Daniel Duffy]

I am starting to wonder when I’ll need to learn FPGAs as well. Anyone else feel the same?

This is a very specialised area. In 99.99% of developers it is not needed.
I would focus on C++20 threads, tasks (futures) and see how that goes.

Learning principles versus learning products.

 

[jeremytec]

Hi Pavlos , may I ask from where did you do you degree in Financial Engineering?
Also do you work on FPGAs as a quant developer?

 

[Pavlos Sakoglou]

Hi Pavlos , may I ask from where did you do you degree in Financial Engineering?
Also do you work on FPGAs as a quant developer?

Hi! I actually don't have a degree in FE, I only took the pre-MFE math courses from Baruch College MFE program and then a few stochastic equations/processes classes during my masters in math at CCNY.

My work does not involve FPGA, but mostly implementation and software architecture of quant libraries.

 

[LINKSKI]

1. Work on more personal projects in detail to compensate for luck of experience

What are projects or extra reading I can do If I am I interested in the quant field as a year 12 (grade 11 US) student.

I recently got into trading and have started backtesting (manually) any analysis techniques I could learn, maybe the basics for a project. I'm not too sure?

Thank you for this opportunity and using our time to read my message

 

[Pavlos Sakoglou]

What are projects or extra reading I can do If I am I interested in the quant field as a year 12 (grade 11 US) student.

I recently got into trading and have started backtesting (manually) any analysis techniques I could learn, maybe the basics for a project. I'm not too sure?

Thank you for this opportunity and using our time to read my message

I think you are way too young to focus your education on one field. I am not discouraging you from working on technical projects, but make sure you keep your mind open, study other things as well, such as history, literature, theater, arts, biology, music, psychology, physics, etc. Also try to get into sports and do social things that are fun. There are valuable lessons and knowledge you can ultimately transfer to your career from everything.

Btw I am talking from my personal experience. Early in my life I briefly worked as a motivational speaker, I was in a band, I did racing kayaking competitively, I wrote a series of essays and ultimately a novel during college, and I studied abroad for a semester. I saw myself growing and developing in such a diverse way which gave me an advantage and a unique skill set that I use to my benefit every day.

 

[Doveman]

UNIX/C++ developer with 4+ years of experience in software engineering, research in statistical learning, and in-depth work in pure

 

[hungrykid]

hey friend. I have some questions if you dont mind me asking! btw english is not my first language so I'm sorry if I have some errors.

To get some context, Im still studing, a business degree but specializing in finance. At the same time that I'm doing a minor in advanced data science (and unfortunatelly, everything with python and only python lol).
But the problem is that the minor is from another faculty of my uni so I really can't "apply" any of my knowledge or what I'm learning in python, into the finance world. And I'm "almost" already working in a bank as a financial risk analyst so I would not have much time to study. That's why I'm trying to focus on what could be useful to get into the quant area without wasting that much time. Considering that, do you really think I should learn c++ ? or trying to get more knowledge in another area? maybe master python first? maybe a specific degree? what about the CFA? any advice????
anything would be really appreciated man! Because I'm really excited with this career path. thanks in advance friend

 

[dgpo]

Hello!

Sorry if this is a bit long!

I currently studying Electrical and Electronic Engineering at a semi-target UK Uni (think Warwick, Manchester, Durham), entering 3rd year, predicted a first class with first class grades in classes like Linear Algebra, Vector Calculus, Differential Equations, Signal Processing and C/C++, and will be taking classes in PDEs and Numerical Analysis. I will also have a ML based 3rd year project (proposed project is on using time series analysis to develop a means reverting strategy for pairs trading so will get more experience). Furthermore I just finished a SWE internship at a top tier investment bank (think GS, JP or MS), with a return offer for another internship.

I am interested in quant finance, as I believe I have the ability and desire to further my maths skills, as well as develop my programming skills and learn more about the markets. I had some exposure to the markets during my internship and I found it absolutely thrilling. The problem is, the best of the best master's courses (i.e. Maths and Finance @ Imperial or MCF at Oxford and basically all the Maths x Finance Masters) only explicitly take maths students, and I'm not sure I'd want to waste my time applying to them as I am almost certain I'll get rejected. I am therefore looking at the following courses that I think have the right blend of math, finance and CS/ML classes (I'll list them in order of my preference):

1. MSc Computational Finance @ UCL
2. MSc Financial Technology/MSc Risk Management and Financial Engineering @ imperial
3. MSc Financial Technology @ UCL (bear in mind the FT and CF courses share almost entirely different core modules, but have more or less the same optional ones)
4. MSc Advanced Computing @ Imperial (I like this because I can choose all my modules, and they are good if I decide to go down the more Quant dev/Algo dev route which is something I am considering, given my background)
5. MSc Computational Finance @ KCL
I am also considering MSc Computational Mathematical Finance at Edinburgh and (if all else fails), perhaps MSc Quantitative Finance from Manchester.

What courses would you recommend I add/remove from there? I'm quite keen on the first 2 and the 4th one, first choice being the UCL CompFin due to the compulsory placement.
Imperial is where the indecision is, my slight preference is FinTech because (in terms of core modules) they still have a good number of maths/stats classes, but perhaps more application and more general skills that would serve me well if I decide to go start my own company for example. I also really like the Financial Econometrics in R/Python module, which isn't on the RMFE course. My only concern is whether Imperial FinTech won’t look as good for quant on my CV/LinkedIn as RMFE, as the mathematical rigour is definitely there, but not as much as RMFE. Then again, my concern with RMFE is that it seems pretty specialist and I might narrow myself down to more middle office quant/risk management positions. I'm stuck here!

Would love to get some feedback!

Thanks a lot!