Module 1: Python for Finance

THE LINGUA FRANCA OF DATA SCIENCE, MACHINE LEARNING, AND FINANCE

• Implement the Kalman, extended Kalman, unscented Kalman, and particle filters in Python and compare their performance on generated data
• Implement a European vanilla option pricer with AAD using the JAX library
• Implement an optimal time series forecaster in Python using existing libraries such as xgBoost

Module 2: Databases in finance – Kdb+/q

AN UNRIVALED TOOL FOR BIG DATA AND HIGH-FREQUENCY DATA

• Implement real-time Expected Shortfall in kdb+/q and visualise it for a fictional portfolio using KX Dashboards
• Build a monitoring and visualisation system for kdb+tick triplets
• Extend the quantQ kdb+/q library to support transformers
• Compare the performance of linear algebra in native kdb+/q and when interfacing with C++ implementations
• Write a generic time series visualisation suite using jQuery on top of kdb+/q

Module 3: C++ fundamentals with use cases from finance

PERFORMANCE, PRODUCTION STABILITY, AND PORTABILITY

• Implement an exotics pricing engine in C++ (you may extend the one developed by Mark Joshi in his “Design Patterns” book)
• Implement a value at risk (VaR) calculator in C++
• Implement an xVA calculation system for linear and nonlinear products in C++

Module 4: Data Structures and Algorithms in C++

THE CORE OF COMPUTING

• Achieve the best possible time complexity for multiplication of large matrices in C++
• Build a benchmarking suite to compare the performance of linear algebra routines in Eigen and XTensor
• Implement a library of lock-free data structures and algorithms in C++ along the lines of STL

Module 5: Designing algorithmic trading applications

STATE-OF-THE-ART CONTENT

• Built a trading strategy backtesting environment on top of PyFolio
• Build low latency middleware for high frequency trading utilising both UDP and TCP
• Port the LMAX disruptor to C++