Dolat Capital Quantitative Analyst / High Frequency Trading Interview Experience

Round 1: General Interview – 27th April, 2020

I knew machine learning before hand and had idea of markets or how they work. So, I had a talk with HR on 4th April for sometimes so that I could study about markets and high frequency trading. I took help from YouTube and Zerodha Varsity for getting familiar from these terms. In 1st round the panel consisted of 2 people and they started asking my projects and work experience in CV. Followed they asked me about LSTM, CNN and how they function also about general machine learning concepts as what is AUC-ROC, what is gradient how does it work, changing alpha parameter. After some more generic machine learning questions they framed questions on projects by changing criteria. After they gave me puzzle i.e. Suppose there are 2 ants on 1m long rod and they are moving towards each other and begin from each end of rod at a speed of 1m/hr after collision they change their direction, find the time after which there will be no ant on rods. They made this puzzle tough by placing 4 ants 2 on end 2 at centre, 2 facing each other. And further they placed ‘n’ ants on the rod and task was to calculate the time. What is the chance that your birthday is not same as some another given ‘n’ number of people.

Round 2: Aptitude Round

There were 10 questions based on topic such as PMI (Principle of Mathematical Induction), Trigonometry, Quadratic Equation, Integrals, 5 puzzles among which famous gold chain cutting puzzle, lift puzzle, permutation and combination around round table. Time given was 1 hour and can be given more if requested.

Round 3 : Coding Test

A software engineer took this round. He asked me basic questions about C and OOP’s concept of C++. Though, my preferred language was Python so, later questions shifted to that, but you need to have a basic idea of C/C++. Then asked me about Data Structure like dictionary, list, tuple, stack etc. and difference between them. He gave me question where you had to write a function where you had to take last 3 LTP (Latest Trading Price) and return the average of it. Designing an execute() function which would do something else. The crux of this round was which data structure would you use and how would you use OOP’s concept to it. He asked about Superclass. 

Round 4 : Final Round – 4th July, 2020

The HR will give you 2 books to read and you will have finish them and revert in 7-14 days. Books : High Frequency Trading by Irene Aldridge, Time Series by Tsay. The final round was taken by a panel of 6-7 members from every domain. They will sum up by beginning with your cv and asking about deep learning (Sequential series, time series). How will you GRU, LSTM in market for an equity, what factors will be you consider apart from LTP while looking at an equity while applying sequential series. How will you use CNN , filter based question, basically checking how good you are with Convolution Neural Network. Then questions from Statistics, mean, median, mode, difference between volatility and liquidity, they will try their to confuse you. When to use z-test, t-test. Questions from books, about Order Book, what are different types of orders. what is candle stick patterns, what is doji how many types of doji, your personal trading experience if any. I had so they asked me about candle stick pattern. They will put forth various case study like what will happen if a very big market order comes in , will the results vary if order is limit order of marketable limit order. Some basic questions on moving average, RSI etc. Finally I was asked question on probability that what is the minimum number of times a dice is to rolled to get 2 consecutive 6’s. 

Note : This is based on purely what I can recall, there were many other questions I faced, if I get reminded of them, I will share it with you. 

P.S. I was offered the role.