Anomaly Detection in High-Performance Trading SystemsJanuary 1, 2022
Tradeweb Computer Science, 2021–22
Liaison(s): Stefan Kutko and Jeremy Jess ’20
Advisor(s): Elizabeth Sweedyk
Students(s): Wenxuan Zhang (PM), Allen Wu, Nicolas Perez Vergel, Nick Tan, Nick Dazell
Tradeweb is the leading trading platform for fixed-income products, derivatives and exchange-traded funds. Tradeweb would like to improve client experience on their platform by minimizing trade latency. The team is testing several machine learning models to predict anomalies in a variety of metrics that are correlated with trade latency. The system will alert Tradeweb of potential latency spikes, so they can respond quickly to resolve problems.