Paramartha Sengupta

Data Scientist

ZS Associates•  October 2021 - May 2023

• Patient Identification: Developed a semi-supervised ML algorithm (PU learning) to identify patients requiring critical healthcare treatment based on their medical history. This model proved effective in identifying 60% patients in urgent need of treatment, compared to a baseline of 25% • Web Usage Optimization: Achieved ~17.2% increase in overall website utility by developing a causal relationship model that establishes the correlation between user interaction on a website to product purchase. Recommendations for website enhancement were made based on this model • Data Preparation Tool: Designed and developed an end-to-end application hosted on Streamlit that provides augmented insights on patients’ journeys and milestones using raw claims data as inputs. Implemented a GPT-3 module API that enables users to control the application using plain English text, without requiring any coding experience • HCP Targeting Model: Currently engaged in the development of a model that predicts the most suitable adopters of a specific drug in the market, utilizing cross-matrices of their potential and propensity Skills: NumPy · Data Modeling · Communication · Computer Science · Machine Learning Algorithms · Keras · TensorFlow · PyTorch · Statistical Modeling · Streamlit · Machine Learning · Data Science · Analytical Skills · Research · Statistics · SQL · Python (Programming Language) · Data Analytics · Deep Learning · Statistical Data Analysis · Supervised Learning · Unsupervised Learning

Data Science Mentor

Kaggle•  December 2022 - March 2023

- Provided mentorship and guidance to mentees on the implementation of ML and Data Science concepts at an Industry level, resulting in the completion of their first end-to-end ML project within three months Link - Developed a feature evaluation module that enables the assessment and evaluation of features’ influence on model’s learning - Designed and developed a customized food recommendation engine that leverages user preference and location choice to predict the best restaurants and dishes

Data Science Analyst

Novartis•  August 2019 - October 2021

Data Modelling Pipeline Development: Contributed to the development of a comprehensive data processing and modelling pipeline that improved global clinical trials monitoring, and designed and implemented a dashboard for continuous performance tracking Field Distance Optimizer: Developed a field distance optimization algorithm for United States and China, which resulted in a 27.1% and 18.7% reduction in overall travel distance for monitoring purposes in the respective countries Risk Analysis: Developed a risk estimation indicator that facilitates the identification of the most promising sites for conducting clinical trials, resulting in ~24% optimization of global monitoring activities Skills: Visualization · Communication · Microsoft Excel · SQL · Python (Programming Language)

NIT, Durgapur (National Institute of Technology)

Chemical Engineering, B.Tech•  August 2015 - April 2019

- Secured a CGPA of 8.98/10.00 - Stood as the Departmental Silver Medalist for the Graduation year 2019