• Developed complex, large-scale SQL-queries involving 2,000,000+ records of end-user platform-activity data using Snowflake, creating customer-profile table for 4+ predictive ML models serving 10,000+ users weekly.• Implemented reliable CI/CD pipelines to handle, process, and deliver ~360MB of ML model-ready data using Pandas, Kubernetes, and AWS (EC2, KMS), delivering 2,600,000+ personalized-recommendations per year.• Executed feature-engineering process to aggregate data, detect patterns, and assign weights for data-model-ingestion using Python and Pandas, increasing relevance and personalization by 4% CVR.• Engineered robust FastAPI Microservice to launch MVP recs-system from 0-1 using Python and Pydantic, generating executive buy-in, shifting company priorities, and repurposing $510,000+ in technical resources.• Constructed a predictive-carrier-sourcing library that searched attributes to produce carrier-matches for 4+ ML-models (30,000+ volume) using Python, helping establish core-platform service for 12+ technical staff. • Defined CI/CD best practices, technical project scope, stable sprint cycles, code reviews, and meeting schedules using Jira and Gitlab, adding structure and clarity for 4+ product leaders and 5+ technical staff.

• Led end-to-end engineering process for enterprise production-improvement Product Suite with data mining and automation using R, Tableau, and SQL, making platform-deep insights accessible for 1900+ employees.• Managed CI/CD best practices, technical project scope, code reviews, and senior stakeholder comms for production-operations data science team using Jira, adding structure and clarity for technical staff of ~16.• Architected full-stack, cost-estimation tool to help secure 12+ production-deals per year ($100M+ TCV per deal) using R, Shiny, SAP HANA, and Domino, eliminating $4,000,000+ worth of manual-estimation cost.• Built fault-detection explainable model containing 865+ records and 150+ features using LIME, h2o GBM, and Dimensionality Reduction, helping 74+ QA staff gather evidence of component-failures (97% accuracy).• Delivered productivity-tool for 5+ work-centers with 16+ machines using R, RESTful APIs, IoT Data, and SAP HANA, enabling 14+ operators to schedule and prioritize 69,120+ hours of component-production tasks.• Deployed internal-application that aggregates 30+ company-wide dashboards for Production & Operations using JavaScript and .NET, helping ~20 execs gain insight on 50+ KPIs for improved data-driven decisions.• Created cyclical-directional graph with 14+ data-attributes like stock-shortage, QA issues, and backlogs using JavaScript, .NET, and D3, visualizing 20,000+ parts within Bill of Materials with critical-path insights.• Produced internal-tool to synthesize 62,000,000+ rows of workload and performance data across ~10 MFC sites using Tableau, helping ~6 executives and 18+ supervisors maximize through-put for 12,000+ operations.• Updated SQL Server Management Studios (SSMS) to move core-platform applications to cloud environments using Domino, refining infrastructure-wide SQL procedures and increasing data-propagation speeds by 27%.

• Gathered ~7.29MB of agricultural data across 16-acres of farmland calibrating Campbell Systems sensors with Python libraries, helping assess 330,000+ crops for viability including 10+ species of sorghum and corn.• Compiled 86,700+ entries of climate-data from 5+ weather stations across Texas (662-mile radius) to build logging-application with JavaScript and .NET, helping study environment-effects on crop-yield in real-time.