Ph.D. student · University of Connecticut

Systems for AI that hold up under real workloads.

I work on machine learning systems, with interests in LLM serving, hybrid retrieval, secure inference, and trustworthy AI infrastructure.

My goal is to build systems that are not only fast, but also robust, measurable, and useful in realistic deployment settings.

Contact me View CV

8+ publications across ML systems, security, and AI infrastructure

4 years of production backend and AI systems experience at Baidu

Now working on LLM serving, hybrid retrieval, and trustworthy AI systems

LLM serving

Efficient inference, KV-cache management, MoE serving, and verification-aware runtime systems.

Coordinating SQL, vector search, filtering, and execution policy under real workload constraints.

Secure inference, attestation, and recoverability-aware infrastructure for dependable AI deployment.