Krishna Jillelamudi

Designed and deployed end-to-end MLOps pipelines (Jenkins, Airflow, GCP) supporting 20+ ML models, reducing deployment time by 40% and improving system resilience across environments.

Led A/B testing and incremental rollout of personalization models using KPI monitoring, driving a 12% increase in model-driven user engagement.

Built SHAP-integrated observability layers and implemented post-deployment XAI workflows, reducing model investigation time by 35% during incident response.

Processed and analyzed terabyte-scale customer behavior data using PySpark on Dataproc, optimizing Spark job performance by 50% in high-traffic pipelines.

Wrote complex, production-grade SQL for customer segmentation and churn feature engineering, accelerating analytics turnaround by 3x for cross-functional data consumers.

Co-led research and course development for CSCI-B455: Principles of Machine Learning, teaching 80+ undergraduate students in advanced ML concepts.

Built Tableau dashboards with ArcGIS integration for 900+ civic organizations, informing infrastructure equity decisions in Indiana municipalities.

Led network analysis of $100M+ grant flows for nonprofit impact evaluation, identifying state-level bias and influencing three academic funding proposals.

Built Azure Data Factory flows for TB-scale mainframe data migration, cutting load window by 6 hours and achieving a 4-hour nightly SLA.

Delivered HIPAA-compliant data orchestration workflows for claims processing, driving a 40% operational cost reduction post-migration.

Facilitated multi-team collaboration during mainframe-to-cloud modernization, ensuring continuity and secure data lineage tracking.