Mid-level Data Scientist specializing in fraud detection and healthcare ML
North Carolina, USAData Analyst4 years experienceMid-LevelFinancial ServicesBankingHealthcare
ScreenedIdentity Verified
Connect with Kavya
Kavya already has a relationship with Reval, so a warm intro from us gets a much better response than cold outreach.
Recommended
Already have an account?
About
Applied NLP/ML in healthcare and financial services, including fine-tuning BERT on unstructured EHR text and building embedding-based similarity search for clinical concepts. Also redesigned a Wells Fargo fraud detection data pipeline using modular Python + AWS Glue/Step Functions, cutting runtime ~40% with improved monitoring and reliability.
Experience
Data AnalystWells Fargo
Principal Data AnalystNextGen Healthcare
Education
University of North Carolina Charlottemaster, Data Science and Business Analytics (2024)
Key Strengths
Improved clinical text classification accuracy ~20% by fine-tuning BERT on EHR data
Built domain-specific preprocessing/tokenization (custom vocab, abbreviation standardization, dataset balancing) to reduce noise and boost model performance
Entity resolution across patient/provider systems using fuzzy matching + manual validation/tuning
Improved semantic linking/search relevance ~25% via domain fine-tuned embeddings (e.g., linking 'heart attack' and 'myocardial infarction')
Designed production-grade Python data workflows with Airflow + AWS Glue, logging/monitoring in CloudWatch, and Git versioning
Reduced fraud detection pipeline runtime ~40% at Wells Fargo by modularizing AWS Glue workflow, enabling parallel tasks and reusable components
Moved models from notebook experimentation to production using AWS SageMaker pipelines with reproducible tracking/versioning
Discover more candidates like Kavya
Search across thousands of pre-screened, high-quality, high-intent candidates on Reval.