Hello everyone,
I received two offers, one as a Machine Learning Quant at Citadel Securities focusing mainly on options, and one as a synthetic products quant at Goldman Sachs S&T. Goldman Sachs made it clear that it would be a role focusing a lot on machine learning. I am really interested by Machine Learning and my issue is, although I would love working with this team at GS (I've already met them), but I think I have much more to learn at Citadel (they seem more advanced on the issue). At the same time, I am affraid of the aggressive culture at Citadel (I don't want to get laid off after a few months) and I think it would be easier for me to make a whole career staying at GS on the long run than at Citadel (especially since I have probably more to add at GS). Citadel would pay more (litterally twice as much), which is not negligeable at this point, but I'd rather bet on the future than solely on present comp.
It is a bit of a difficult choice for me because I've already said yes to GS, but I'm still torned as Citadel came back to me recently.

I am not expecting a clear answer, but maybe insiders could enlighten me a bit.

Comments (12)


I don't work in machine learning or anything like that but I do work on a trade floor. Absolutely take the Citadel job. If you want to excel in this business you can't have the timid mindset of being scared of an aggressive culture. You should look at that type of culture as an opportunity to absolutely kill it, makes tons of money, and learn more than anyone else in your peer group. Look at the potential success, not the potential downside. Not one famous trader or CEO you have ever heard about was scared to fail at anything they did. They saw difficult opportunities as chances to make a name for themselves, not fail.

Secondly, the money seems to be better at Citadel as well so that seems to be the icing on the cake.

Learn More

814 questions across 165 hedge funds. 10+ Sample Pitches (Short and Long) with Template Files. The WSO Hedge Fund Interview Prep Course has everything you'll ever need to land the most coveted jobs on the buyside. Learn more.


Although turning down an accepted offer is obviously not good, I don't think it'd burn all bridges. If you want to stay within ML, Citadel is the obvious choice. Not sure if this thread is just to brag.


It is absolutely not to brag, I couldn't care less about that.
I didn't know Citadel was such an obvious choice according to you.


Citadel is at least on the same level as GS when it comes to machine learning (naturally, it'd depend also on the specifics of the job). Additionally, with Citadel on your CV - even in worst case (i.e. a potential layoff) - you'll basically have all doors open. Overall, IMO GS is very decent whereas Citadel is quite unique.


Obviously Citadel is a much better place to be. The only shortcoming is that it seems to be a relatively silo-ed place/ everyone-keeping-own-IP style. However, at current stage, I don't think an IB job will even allow your ML result (probably they would allow liner regression/logit regression and I think it should be useful already) to be used to trading at all (funky model review + endless bureaucracy and lack of infrastructure in case you have a time consuming evaluation process).

Joining GS will teach you how to do basic analysis, which I bet you could pick up in months. You should be able to know what your colleague is doing. Except for that, I think a big pro is that being in GS, you are likely to interact with client's trade where knowing how clients trade ( strategy/idea or simply execution information) is a great knowledge to learn and being able to play around with those data and knowledge will help a lot in alpha research in the future (in a place like Citadel).

Citadel will have tons of much better quants than GS, just think about the best quants in GS will eventually either join a quant fund ( Citadel's name stay on the top of the list) or an HFT firm (Citadel's name stay on the top of the list, again...). However, joining fresh from undergrad/grad school may or may not be a good choice to be in since the mentorship from others stays within the thought process and skills needed (which I think is the most important thing) but you will have to find your own idea and have little idea what others have done. You probably have no way to know the whole picture of the business (again, good for the firm, bad for individuals) . Since you have already got an offer from Citadel, and have said yes to GS, starting in GS for 1 year or 2 to learn the basic stuff and then join Citadel makes more sense to me. (or other hedge funds and there are a ton of them out there still, like Two Sigma/ DE Shaw/ Man, or smaller ones like Cubist, they are all very good quant shops and with a resume that has got into Citadel + several years in GS, I don't think you will have trouble getting into one of them)


Thank you a lot for this detailed answer. It will really help me take a decision. I think you are right, if I really don't like the work at GS, I can always move to another firm. I will probably stick with GS for now.


You definitely shouldn't renege on GS since you've accepted the offer, but my take is that Citadel would've been the choice here -- Citadel is an amazing shop, and I get the sense (from personal anecdote, and from reading a bit about their selection process and strategy) that the quants there are the brightest in the industry. Definitely keep in contact with the people you came in contact with at Citadel! If after two years you do want to circle back around, I'm sure those people will be of great assistance.


Yeah don't work in quant field but in general this advice of not taking the "risky" option to start your career making 2x the money and working with way smarter people is idiotic. You can't "always move to Citadel in a year if you don't like GS". Fucking idiotic advice.


1-Click to Unlock All Comments - 100% FREE

Why do I need to be signed in?
WSO is a knowledge-sharing community that depends on everyone being able to pitch in when they know something.
+ Bonus: 6 Free Financial Modeling Lessons with 1-Click Signup ($199 value)