Hank Marquardt

Write, review, and compare code solutions used to train and evaluate large language models on software-engineering tasks. Evaluate correctness, reasoning quality, maintainability, edge-case handling, and provide feedback to improve model reliability.

Analyze audio sentiment, speech patterns, and conversational intent to produce training and evaluation data for AI systems. Perform Turing-style evaluations and detailed prosodic annotation while applying guidelines consistently and documenting quality issues.

Built and operated AWS-based data workflows integrating vendor, API, file, and point-of-sale data into Redshift and S3. Designed SQL data models and reusable pipelines and coordinated data migrations for acquired companies to support BI and operational decision-making.

Led a team of 6+ developers delivering client-facing digital engagement platforms, technical delivery, and production support. Designed event registration and reporting/dashboard solutions, translated client needs into technical roadmaps, and improved delivery and operations through automation and troubleshooting.

Maintained and enhanced ETL pipelines for marketing and performance dashboards using Microsoft SQL Server, Python, and Ruby. Built feature attribution and transformation logic for digital/media performance analysis and partnered with analysts to diagnose data-quality issues.

Developed and supported e-commerce, order-entry, fulfillment, and internal business applications in a long-term enterprise environment. Built APIs and integrations connecting Amazon and eBay storefronts, automated operational workflows, and helped modernize legacy processes while maintaining production support.