Quant skills can be divided into three categories:
Mathematics
Computer Science
Finance
95% of your time should be spent building competency in math/CS with 5% of your time allocated to a baseline understanding of derivatives pricing, fixed income mathematics, portfolio theory, and knowledge of the capital structure. Once you get your first job, you will tailor your finance learning to the specific strategies employed by your firm.
When learning, it is important to differentiate between knowledge and skills.
Mathematics and programming are skills
Finance is mostly knowledge (in the context of what you need to know for interviews)
Knowledge represents that which can be learned. Skills represent that which must be trained. I can teach you the capital asset pricing model in a single lecture. I can’t teach you how to solve extremely difficult algorithm problems, design an elegant software system, or find an illusive alpha. These skills can only be acquired through hours of deliberate practice.
A quant is a mental athlete and interviews are tryouts. Your mathematics and programming skills are analogous to the physical strength of a wrestler. Being physically strong won’t make you a champion, but being weak will disqualify you entirely.
Many candidates make the mistake of over allocating time to building knowledge and under-allocating time to practicing skills. This is natural because the hard skills are not specific to quantitative finance so it can feel like you aren’t doing much to advance your career goals.
Mathematics Skills
Probability
Linear Algebra
You can carry yourself through most interviews with probability and linear algebra. My advice for getting good at probability is to crack open the Sheldon Ross textbook and just solve every question. The textbook contains lots of solutions to problems and plenty of conceptual proof questions. You could spend 6 months living inside of that book and emerge a master of probability.
For linear algebra, there is the classic Gilbert Strang book. This book is a classic tome on linear algebra and is generally thought to be one of the best texts on the topic. Linear algebra is a finicky topic because of the marriage between its applications and theoretical components. A significant amount of time should be devoted to understanding the major results of linear algebra to build a robust intuition around common data science problems.
Computer Science
Object Oriented Design
Algorithms and Data Structures
Object oriented design is best learned through a formal university course. If you do not have the option to take a formal class, try finding one on coursera. Alternatively, you can buy this scrappy project based OOD book in python.
The algorithms and data structures knowledge tested in quant interviews does not surpass that of an undergraduate university course. I recommend starting with an algorithms course through an online learning platform like MIT open courseware or coursera. After completing the course, head over to leeetcode and solve a few hundred algorithm questions. Get to a point where you can tackle most medium-difficulty problems.
Finance
The finance books attached are hybrids between textbooks and industry primers. The goal with reading these books is to make you conversant in the field so that you can show interviewers that you are serious and passionate about working at a hedge fund. This will differentiate you from the typical STEM student who is undecided between quant vs FAANG.
Advances in Active Portfolio Management, Richard Grinold and Ronald Kahn
Quantitative Hedge Funds: Systematic, AI, ESG and Quantemental, Richard D Bateson
What’s Next
After you have developed core competencies in probability, linear algebra, object oriented design, and algorithms your next step is to practice real interview questions.
Interview Prep Resources
This Blog (Weekly interview puzzles right to your inbox!)
Cut The Knot [Probability brainteaser book]
Advice
Build a regular practice of solving a few interview questions every day. It is much easier to turn interview prep into a habit than to grind it out in 8 hour marathons. Additionally, if you are able to solve a question, tweak the parameters of the question and see if you still know how to solve it. This is useful for making sure that you’re building your problem solving skillset rather than memorizing questions.