Data Scientist
I build machine learning systems, data products, and research tools that turn complex information into decisions people can inspect and trust.
I am a data scientist with 6+ years of experience across healthcare, financial services, and analytics. I work across the ML lifecycle, from data quality and feature engineering through evaluation, deployment, monitoring, and stakeholder-facing tools.
I build end-to-end projects around real data, explicit evaluation, and usable interfaces.
I built a citation-grounded Canadian drug information assistant over official Health Canada and Ontario formulary sources.
15,562 DIN-level records, 1,327 traceable chunks, 87.5% Recall@3, 100% citation validity, and 95% behavior accuracy.
I built a text-to-SQL research assistant for inspecting historical strategy behavior without handing raw SQL access to the model.
4.1M market records, 21 read-only MCP tools, 124 passing tests, 0% tool hallucination, and 0 safety violations.
I built a leakage-safe research and execution platform for trade filtering and position sizing across five futures markets.
Eight model families, walk-forward and Monte Carlo validation, Dockerized multi-broker execution, and 83 to 177 tests passing across 10 clean release runs.
I present these as research and analytics applications. My trading workbench is not an investment advisory product and does not predict future returns.
Aug 2023 to Jan 2026
Claritev
At Claritev, I streamlined ProPricer claims analytics across 2M+ monthly claims, built adaptive risk thresholds supporting up to $22M in potential annual savings per client, automated Python workflows, optimized SQL pipelines, and operated CERES, an XGBoost-based benefit-utilization platform.
Aug 2022 to Dec 2022
Connecticut Information Technology Institute, UConn
I built Selenium workflows for 2,000+ clients, developed time-series forecasting models, ran landing-page A/B tests, and delivered Power BI, Tableau, and Streamlit dashboards.
May 2022 to Jul 2022
American Century Investments
I forecasted ESG controversy sentiment with NLP, deployed Random Forest and BERT models on AWS at 87% accuracy, and supported KPI analysis, ETL, and ANOVA testing.
Aug 2021 to May 2022
Connecticut Information Technology Institute, UConn
I designed predictive time-series models, built an 87% accuracy Random Forest classifier on Azure ML, delivered Tableau and Streamlit dashboards, and supported data-driven web design and digital transformation.
May 2017 to Jul 2021
Alliant Infotech LTD
I built PCA-based customer segmentation, predictive models, dashboards, ETL pipelines, and Excel reporting automation, cutting campaign costs and processing time.
Open the two-page resume preview or download the PDF.
If you want to discuss applied ML, data products, or a research workflow that needs more rigor, reach me directly.
kshp95@gmail.com