- I lead initiatives through discovery, scoping, and delivery working closely with senior leadership and business partners, actively seeking new opportunities to utilize cutting-edge AI/ML technologies for efficiency gains and improved user experience.- Built NLP models to extract key user-centric insights about targeted product groups and service journeys.(Technologies: aspect extraction, topic modeling, sentiment analysis, and clustering)- Built an attribution model to quantify performance drivers for 40+ customer journeys; Built a complimentary model to predict user satisfaction over time(Technologies: regression and classification modeling, time series forecasting, explainable AI tools)- I coach individuals on technical and business competencies, lead various learning initiatives for existing and aspiring data scientists, and support data science recruiting.

Build data models to predict insurance and healthcare costs to users while accounting for nuances in the data like extreme high-cardinality, inaccuracies and incompleteness. Perform advanced analytics centered around conversions, user behavior patterns and user experience using natural language processing techniques and classification algorithms. Leverage statistical tests of significance for experiment design, test and control targeting, data balance and experiment outcome analysis. As part of a multi-disciplinary team in a fast-paced environment, I work closely with engineers to deploy models into DevOps pipelines, participate in weekly user interviews and research synthesis, and work with UX designers and product managers to review user feedback in order to identify opportunities for model iterations. Successfully adopted agile for data science including working with product managers to write stories and acceptance criteria for better stakeholder visibility as well as to keep data science efforts aligned with product portfolio goals

Collected and analyzed business requirements, translated to data requirements to create analytical, predictive and prescriptive solutions. Automated piping of structured and unstructured data from multiple sources, assessed data source and verified data integrity. Developed regression-based tools to generate test users modeled after real users’ demography. Performed principal component analysis and built classification models to forecast primary user activity. Developed text mining and classification tools to automate manual review of historical requirements and results. Visualized findings, presented results and recommendations to stakeholders and mined critical insights.

Worked as a consultant analyst for global clients. Performed requirements review, data acquisition, project planning and user acceptance. Processed & cleansed data; performed missing data imputation & feature engineering to standardize data obtained from multiple sources for predictive modeling. Modularized client’s application framework to increase efficiency and recommended business features to reduce churn.

Worked with global clients and engaged in several projects for to improve user experience and application performance. Collected/reviewed client requirements & created design documents. Cleaned & processed datasets; performed exploratory data analysis. Developed & implemented demography based profiling tool for targeted advertisement. Collaborated with several teams to improve decision support system to increase efficiency and reduce costs.