Battu Alekya

Data Scientist @Wells Fargo

US

Designed and deployed financial forecasting models leveraging statistical and machine learning techniques to predict stock market trends and economic indicators, enabling improved strategic planning and risk management for enterprise clients. Developed and product ionized time-series and ensemble-based financial forecasting models (e.g, ARIMA, XGBoost, LSTM) to anticipate stock price movements and macroeconomic indicators, leading to a 28% improvement in portfolio risk mitigation for institutional clients.Performed deep exploratory data analysis (EDA) using Python libraries such as pandas, seaborn, and matplotlib, uncovering key temporal and categorical trends that informed model feature engineering strategies and reduced model error by 15%.Built and maintained a modular MLOps infrastructure using MLflow, DVC, and Docker to manage the full ML lifecycle enabling consistent model retraining, reproducibility across environments, and traceable experiment tracking.Implemented end-to-end CI/CD pipelines using Jenkins and GitHub Actions to automate testing, model packaging, and containerized deployment on AWS SageMaker, reducing deployment time by 70% and eliminating manual integration steps.Architected robust ETL pipelines using Python (airflow, pandas) and SQL to handle multi-source financial data ingestion, transformation, and validation reducing preprocessing latency by 50% and improving downstream model accuracy via cleaner data.Deployed state-of-the-art NLP workflows using transformer-based models (e.g, BERT, RoBERTa) and libraries such as SpaCy and Hugging Face Transformers to perform topic modeling and sentiment classification on financial documents, increasing predictive precision by 22%.Led statistical modeling initiatives, conducting A/B testing, ANOVA, and linear/multivariate regression to evaluate market interventions and pricing strategies, directly contributing to a 12% increase in campaign ROI.