Technical and project lead on implementation of deep quantile regression models in TensorFlow to identify anomalous data submissions. Results showed double-digit improvement in related KPI.

Led project to develop and implement methodology to cluster financial institutions and highlight concentrations of risk in the financial system using bayesian gaussian mixture model (scikit-learn).

Led project to develop and implement a methodology to rank the effectiveness of data validation rules, using both natural language processing (scikit-learn) and statistical techniques (pandas), highlighting opportunities to drive more efficient operations.

Built NLP models (scikit-learn) to attach metadata tags to documents submitted by supervised firms, resulting in wider adoption of open source technologies at the Bank.

Led project to collect business problems across the Bank that could be solved with machine learning, resulting in pipeline of high-profile projects from the Bank’s largest groups.

Coordinated and led sessions to teach senior policy makers and analysts about machine learning and the latest technologies being used in the FinTech sector, including TensorFlow and DeepLearning.

Implemented two methodologies to benchmark liquidity models used by large and complex supervised institutions, resulting in greater understanding of systemic risk in the financial system.

Developed SQL backend for dashboard to track key performance metrics of operational teams in the Statistics Group.