Tao Wei
Professor of Electrical & Computer Engineering · Clemson University
Leads the NEXT Lab focused on domain-specific accelerators, reconfigurable computing, and applied ML. Guides FastFlowLM kernel strategy and academic collaborations.
Team
Architecture + systems + community
FastFlowLM is a collaboration between academic researchers, software engineers, and community maintainers. The core group includes four PhDs and three B.S. graduates in Electrical and Computer Engineering with deep experience in LLM internals, parallel processing, and architecture-specific software optimization.
What we focus on
-
Kernel research
Co-designing fused attention + MoE operators with AMD.
-
Developer experience
Shipping ergonomic CLI, APIs, and docs inspired by Ollama.
-
Open community
Weekly office hours, demos, and benchmarking nights.
Core collaborators
Hardware + runtime expertise
Ken Qing Yang
Distinguished Engineering Professor · University of Rhode Island
With more than 30 years of experience in computer architecture and parallel processing, he is a serial entrepreneur who has successfully built four high-tech startups rooted in his research innovations—including VeloBit (acquired by Western Digital) and DapuStor (currently in the IPO process).
Zhenyu (Alfred) Xu
Research Assistant Professor · Clemson University
Focused on domain-specific accelerator design, reconfigurable computing, and efficient on-device AI inference. Brings deep experience in hardware–software co-optimization across FPGA, CGRA, and emerging AI accelerator architectures.
Advisors & contributors
Builder network
FastFlowLM thrives because of community engineers who maintain connectors, polish docs, and stress-test nightly builds.
Press & partnerships
Let’s make Ryzen AI shine
Hardware vendors, ISVs, and research labs collaborate with FastFlowLM for demos, co-marketing, and silicon feedback loops. Drop us a line to get started.