Welcome to My Academic Portfolio

👤 About Me

I am a first-year PhD student at University of Illinois Urbana Champaign (UIUC) 🎓 advised by Prof. Minjia Zhang. My research focuses on high performance large-scale LLM training 🚀.

🔬 Research Interests

My research interests span high-performance computing 💻, machine learning systems 🤖, and GPU optimization ⚡. I focus on:

• ⚙️ Large-scale language model training and inference optimization
• 🎯 GPU kernel optimization
• 🏗️ HPC

🎓 Education

Ph.D. in Computer Science 📖 (In Progress)
University of Illinois Urbana-Champaign, 2025-Present
Advisor: Prof. Minjia Zhang

BS in Computer Science 🎯 (2021 - 2025)
University of California, San Diego
Advisor: Leon Bergen, Mohan Paturi, Taylor Berg-Kirkpatrick

📚 Publications

Quiet Feature Learning in Algorithmic Tasks 🧠
P. Naidu, Z. Wang, L. Bergen, R. Paturi
AAAI 2026 Special Track on AI Alignment (Accepted as Oral Presentation) 🎤

The Surprising Soupability of Documents in State Space Models 📄
Y. Jafari*, Z. Wang*, L. Bergen, T. Berg-Kirkpatrick
arXiv preprint arXiv:2505.24033, 2025

Omniwise: Predicting GPU Kernels Performance with LLMs 🔮
Z. Wang, C. Ramos, M. A. Awad, K. Lowery
In submission to U.S. Patent Application.

Preliminary Results of the MLPerf BERT Inference Benchmark on AMD Instinct GPUs 📊
Z. Wang, K. Vu, M. Hodak, A. Mehrotra, F. Gutierrez, K. Smith, G. Seo, et al.
Practice and Experience in Advanced Research Computing 2024: Human Powered Computing, 2024

🏆 Competitions & Awards

Student Cluster Competition (SC24) - Team Co-Lead, MLPerf Benchmark Lead 🥈
SC Conference Series (Mar 2024 - Nov 2024)
Atlanta, Georgia

• 1st Place Among U.S. & Europe Teams, 4th Place Overall
• 1st in MLPerf Benchmark in AMD GPUs (2nd in all GPU groups)
• Designed heterogeneous 3-node system with 2 GPU nodes (1x EPYC Genoa 9534, 4x Instinct MI210) and 1 CPU node (1x EPYC Genoa 9634)
• Supported first open-source multi-node optimized MIGraphX backend for MLPerf SDXL Inference benchmark

Student Cluster Competition (SC23) - Team Co-Lead, MLPerf Benchmark Lead 🥉
SC Conference Series (Mar 2023 - Nov 2023)

• 3rd Place Overall, 1st Among U.S. Teams
• 1st in MLPerf Benchmark
• Designed 3-node enterprise computational cluster with 12 AMD Instinct MI210 GPUs and 6 AMD EPYC 9684x CPUs
• Represented UCSD competing against world’s best HPC teams at Supercomputing Conference 23

💼 Experience

Internships 💡

Performance Engineer (MLPerf Team) ⚡
AMD (May 2025 - Aug 2025)
Manager: Miro Hodak
Santa Clara County, California

• Core contributor to MLPerf Inference 5.1 Llama3 405B Submission
• Core contributor to MLPerf Training Llama 3 8B training benchmark

Research Intern (RAD Profiling Team) 🔬
AMD (Jun 2024 - Sep 2024)
Manager: Keith Lowery
Austin, Texas

• Conducted LLM-guided software optimization research under Keith Lowery Group
• Co-authored two Invention Disclosure Forms (IDFs)
• In process of US Patent Application

  News & Updates