Work experience
---------------------------------------------------------------------------------------------------------------------------------
TrustedShops (Dec 2023 - Present)
Data scientist
Collaborated with marketing to identify leads by scraping competitor and market websites, expanding the potential customer base.
Built customer segmentation models and a churn-risk PoC (customer health score) to support retention
Developed a local LLM-based PoC for automated shop categorization to explore scalable classification solutions.
Blackswan Technologies (Feb 2020 - Nov 2023)
Data science consultant
Developed a schema-matcher to determine semantic attribute type and aggregate information from multiple sources, leveraging TensorFlow and Python.
Utilized transfer learning and Python to create a Generic Attribute Matcher as a part of a comprehensive Entity Matching solution.
Leveraged expertise in TensorFlow and Amazon Web Services (AWS) to drive innovation and improve data-driven decision-making.
Moody's Analytics (June 2018 - Feb 2020)
Associate Director, Data Science
Developed a Recommendation Engine for Courses to recommend courses to students of the finance institute CSI, utilizing user-based collaborative filtering and Python.
Reduced efforts of financial analysts in creating company profiles by categorizing finance news using XGBoost and USE, and Python
Analyzed job market and extracted relevant skills, certifications, and education information from JDs, using customized NER, Streamlit, and Python, to help CSI update its courses.
Altisource Labs (Aug 2016 - June 2018)
Lead Analyst, Data Science
Developed a Document Classification System for the loan origination industry, utilizing XGBoost and R, with the ability to classify over 200 types of documents.
Developed a US House Market Forecast System that analyzes trends in various regions of the US housing market. Utilized data from the US Census over the past 15 years, including factors such as average income and population density. Implemented LSTM and K-Means techniques using Python for data analysis and forecasting.