Bengaluru, Karnataka, India
As a Senior Data Scientist at Citi, I specialize in forecasting, quantitative finance, and the development of AI systems for fraud detection. My work includes designing and deploying advanced AI systems, such as a multi-agent orchestration framework and neural network architectures, to enhance decision-making in financial operations. By contributing to initiatives like reducing page load times and boosting user engagement, I have supported innovative solutions in financial data applications. With a Master's degree in Management from the Indian Institute of Science (IISc), my academic foundation complements my practical experience. My expertise includes building robust forecasting algorithms, implementing neural network models, and utilizing machine learning techniques to address complex financial challenges. I am committed to leveraging data science to drive strategic insights and innovation in the financial sector.
• Architected a multi-agent orchestration framework using LangGraph to automate complex financial inquiries. • Designed and deployed AI systems for advanced fraud detection, reducing fraudulent transactions by 25%. • Owned the end-to-end experimentation pipeline, delivering a 60% reduction in page load time and a 50% boost in user engagement.
At Citi, I played a pivotal role in designing a Neural Network architecture based on Temporal Convolution Networks (TCN) for real-time anomaly detection. This innovative approach allowed for effective monitoring of signals across multiple currency pairs, supporting critical treasury decisions. By implementing state-of-the-art deep learning models, I enabled the identification of statistical arbitrage opportunities, significantly enhancing decision-making processes for global treasury operations.
• Conducted quantitative research analysis in ML, NLP, and financial engineering for liquidity management solutions. • Developed robust forecasting algorithms using RNN and LSTM models, enhancing accuracy in financial predictions. • Delivered a Time Series Clustering framework to identify client behavioral patterns, aiding treasury cash management. • Collaborated with stakeholders to translate complex insights into strategic initiatives for non-technical audiences.