I am on the job market for full-time research scientist/engineer positions. Email for CV if interested!

---

Brief Bio

I am a final-year PhD candidate at the University of Central Florida, advised by Jun Wang. I build efficient, scalable, and fault-tolerant machine learning systems (MLSys) for large-scale foundation models. I am one of the cohorts for ML and System Rising Stars 2025.

08/2025 - 02/2026:	Student researcher at Google. I worked on post-training optimization for Gemini on-device deployment.
05/2025 - 08/2025:	Research Intern at Microsoft Research. I worked on system-driven test-time scaling for reasoning models.
05/2024 - 08/2024:	Research Intern at d-Matrix. I worked on system optimization for diffusion-based video generation models.

---

Selected Publications

• GhostServe: A Lightweight Checkpointing System in the Shadow for Fault-Tolerant LLM Serving Shakya Jayakody*, Youpeng Zhao*†, Chinmay Nehate, Jun Wang MLSys 2026 [PDF] [Code] (*: equal contribution ,  †: project lead)
• MeRino: Entropy-driven Design for Generative Language Models on IoT Devices Youpeng Zhao, Ming Lin, Huadong Tang, Qiang Wu, Jun Wang AAAI 2025 [Arxiv]
• ALISE: Accelerating Large Language Model Serving with Speculative Scheduling Youpeng Zhao, Jun Wang ICCAD 2024 [Arxiv]
• ALISA: Accelerating Large Language Model Inference via Sparsity-Aware KV Caching Youpeng Zhao, Di Wu, Jun Wang ISCA 2024 [Arxiv]

---

Links

• Google Scholar.
  
• LinkedIn.
  
• GitHub.
  

---

Fun Facts

• I used to play some elecctronic keyboard and won a national prize in 2008 🎹
• I played a bit college Esports (R6 Siege) and was the founding member of GT R6 Team in 2020 ⚔️
• My Chinese name is 有朋, which means having friends from all over the world 🤓
• My English name is Kenneth, but people usually call me Ken or Kenny 😎