Shivani Gupta

Objective

• Enable seamless search of Big Data medical research images & documents on new NLM search interface

Responsibilities

• Mapped biomedical research repository attributes to defined DATMM classes
• Implemented JSON LD data structure aligned with RDF schema
• Designed GraphQL queries for enhanced search and generated tableau dashboards for visualizations
• Implemented python based A/B testing to evaluate impact of incremental schema updates on search experience
• Automated repetitive tasks using scripting languages such as Python or R, saving time during the analytical process significantly.
• Utilized advanced querying, visualization and analytics tools to analyze and process complex data sets.
• Applied statistical and algebraic techniques to interpret key points from gathered data.
• Worked with stakeholders to develop quarterly roadmaps based on impact, effort and test coordinations.
• Designed interactive Tableau dashboards and reports, providing real-time business intelligence to stakeholders.

Objective:

• Automated standardization of ambiguous medical research document data submitted to NLM's common data repositories

Responsibilities:

• Translated business requirements into data-driven solutions, providing value-added insights that directly contributed to the project goals.
• Developed custom ETL data pipelines for streamlined data ingestion and processing, improving workflow efficiency & extract data from research documents
• Analyzed large datasets to identify trends and patterns in submitted repository data
• Generated embeddings using Sentence Transformers & loaded them in Vector DB FAISS & metadata in GCP to aid similarity search for documents
• Compiled, cleaned and manipulated data for proper handling
• Coached and mentored junior data scientists on SAS and data mining techniques.
• Identified, measured and recommended improvement strategies for KPIs across business areas.

Objective

• Standardize PubMed portal search input fields based on end user search behavior analytics

Responsibilities

• Programmed complex SQL queries on PubMed portal search logs of 5 years & identified user input pattern across 48 free text input fields
• Enumerated succinct list of input values for 48 search fields to remove free text user input
• Enhanced application functionality by developing custom stored procedures, triggers, and functions using SQL
• Optimized search performance by implementing efficient indexing and query optimization techniques.
• Contributed to project planning by providing accurate time estimates for programming tasks and milestones
• Ensured data integrity with the implementation of comprehensive error handling and validation procedures.
• Collaborated with system architects, design analysts and others to understand program requirements
•