Harold Hersey

Experienced software architect with over 40 years of diverse industry experience, integrating deep systems engineering with cutting-edge AI agent frameworks. Specializes in Rust-based architectures and the Model Context Protocol (MCP) to develop intelligent, context-aware systems. Proven track record in revitalizing legacy systems, constructing scalable infrastructure, and leading global teams to deliver impactful software solutions. Passionate about deploying practical agentic AI solutions and cultivating high-performance machine learning platforms, with extensive expertise in full-stack Java, SQL, and more.

• 10+ years of software development leadership, including globally distributed teams
• Agile coaching (Scrum, Kanban, Kaizen), team building, and cross-functional facilitation
• Solutions architecture, crisis management, budget oversight, and strategic planning
• Vendor/client relationship management across offshore and local stakeholders
• Technical mentorship, career development, and infrastructure team leadership

• Harold is not only a highly skilled developer, he's also a fantastic leader, mentor, and teacher..., Eddie Olivas, Senior Software Engineer
• If I could pick my 'dream team,' Harold would be at the top of the list..., Shad Bolling, Technical Writer
• One of the most gifted architects and team leads I've worked with..., Andrew Kirk, Software Engineer
• A force-multiplier on any software team., Paul Klipp, Product Manager

Building Domain-Specific Environments for Computational Science: A Case Study in Seismic Tomography International Journal of Supercomputing Applications and High Performance Computing, Sep 1997

This paper describes developing a computational environment for marine seismologists studying mid-ocean ridge volcanism through tomographic image analysis. The environment integrated high-performance parallel computing, large data analysis, model visualization, and interactive computation control.

The collaboration benefited both disciplines: geologists made fundamental discoveries about the East Pacific Rise and improved parallel ray tracing algorithms, while computer scientists validated their programming, analysis, and visualization tools—particularly DAQV (parallel program data query) and Viz (visualization programming)—in a real scientific application. The authors detail both the successes and limitations of creating this integrated computational environment for seismic tomography.