Software engineer working at the intersection of high-performance computing, heterogeneous systems, C++, and AI-assisted performance engineering.

I am a researcher at AMD Research and Advanced Development (RAD), where I contribute to runtime systems and tools for GPUs and AI accelerators, and investigate AI-assisted performance engineering.

I earned my Ph.D. from Texas A&M University, researching runtime systems for large-scale parallel computing. Read my dissertation.

• ROCR - Leading the effort to extend the ROCm runtime to support AMD XDNA accelerators alongside GPU compute.

• MLIR-AIE, IRON, Triton-XDNA - Contributing to programming tools for Ryzen™ AI NPUs and AI Engines.

• IntelliKit - Building skills and MCP servers that give AI agents structured access to performance-engineering workflows.

• STAPL - Developed its runtime system (STAPL-RTS), including asynchronous nested parallelism and unified shared- and distributed-memory execution. Built as part of my Ph.D. dissertation under Dr. Lawrence Rauchwerger.

Some of my code ships from @ypapadop-amd - that's my AMD account, and it's where my day-to-day work lands. Browse the merged pull requests

• deferred - Deferred evaluation expressions for modern C++23.

• leafwiki-mcp - An MCP server that lets agents use LeafWiki as persistent memory.

You can find me at:

• website
• LinkedIn
• StackOverflow
• Twitter/X

Code:

• GitHub
• GitLab

Academic:

• ORCID
• dblp
• Google Scholar
• ResearchGate
• Web of Science
• Mathematics Genealogy Project

Other:

• Google Patents