Bryan Armstrong

Designed and built backend services in Python to support distributed application workflows. Developed REST API integrations with standardized error handling and reliable data exchange. Optimized PostgreSQL queries and indexing strategies, reducing API latency by 30%. Implemented schema migrations and database validation practices to improve data consistency and reduce deployment-related database issues by 25%. Built reusable backend components for request validation, authentication, and structured error handling. Strengthened reliability with automated unit/integration tests, reducing recurring production regressions by 28%, and improved observability with structured logging and actionable alerts, reducing troubleshooting time by 35%. Containerized services with Docker and improved CI/CD pipelines through automated build and validation. Integrated external APIs with resilient timeout/retry/failure-handling, reducing integration failures by 24%. Collaborated with frontend engineers on A

Designed and developed large-scale Spark and Python data pipelines processing 150M+ shipment events to support production machine learning for ETA prediction and logistics optimization. Built and deployed XGBoost and scikit-learn models for shipment delay prediction, improving early risk detection by 24% and reducing missed delivery commitments by 13% through feature engineering and model optimization. Architected scalable ETL pipelines using Apache Spark, Hive, Airflow, SQL, and AWS, reducing nightly processing time by 42% while improving data quality, lineage, and reliability. Developed forecasting models for package volume and workforce planning, improving accuracy by 15% and reducing operational overtime costs by 12%. Optimized distributed Spark jobs and enterprise SQL workloads, reducing dashboard latency by 45% and lowering cluster resource consumption. Built interactive Tableau/Power BI dashboards, reducing manual reporting effort by 70%. Developed predictive maintenance models

Designed and maintained scalable ETL pipelines for healthcare claims, policy, customer, and wellness data using Python, SQL, Airflow, Spark, Hive, and AWS. Built distributed batch processing pipelines transforming large datasets into analytics-ready outputs for predictive modeling and business intelligence. Developed REST API ingestion pipelines integrating internal enterprise systems and third-party healthcare platforms. Designed and optimized cloud data warehouse solutions using Amazon Redshift and PostgreSQL for high-performance analytics. Improved pipeline reliability by 29% with automated monitoring, data validation, alerting, and fault recovery. Reduced end-to-end data latency by 33% via Spark workload optimization, Airflow scheduling improvements, incremental processing, and distributed ETL architecture. Implemented data quality frameworks including schema validation, anomaly detection, duplicate elimination, and reconciliation. Delivered HIPAA-compliant solutions in partnership