Experience

Industry internships, research, education, teaching, and service.

May 2026 – Aug 2026
AI Research Scientist Intern
Arm · Seattle, WA
  • Built a runtime for attention–FFN disaggregation (AFD) in LLM inference, running attention and feed-forward layers as separate serving stages.
  • Implemented context-length-aware request routing and expert routing to handle workload and mixture-of-experts imbalance.
  • Evaluated on models of up to 2.4T parameters across 36 GB200 GPUs in an NVL72 system: up to 67% higher token throughput and up to 28% lower p99 inter-token latency than vLLM.
LLM inferencedisaggregated servingMoEGB200 NVL72vLLM
Aug 2025 – Dec 2025
Applied Scientist Intern
Amazon Web Services · Santa Clara, CA
  • Quantified LLM serving optimizations (prefill/decode disaggregation, continuous batching, prefix caching) and tuned vLLM runtime parameters for deployments.
  • Built KernelSight-LM, a kernel-level LLM inference simulator that combines kernel-latency, communication, and host-overhead models in a discrete-event scheduler; 2.7% median throughput-prediction error (target-measured tier) across six model families. It extends the open-source Vidur serving simulator.
GPU performance modelingsimulationvLLM
Sep 2023 – present
PhD Researcher
CAAD Lab, Boston University · Boston, MA
  • Systems for large language models with Prof. Martin Herbordt: programmable switches and SmartNICs for distributed training and inference, FPGA-accelerated secure ML, and AI for performance engineering.
  • Papers at ACM ICS, HEART, and IEEE HPEC, including the HPEC 2025 Outstanding Student Paper Award.
FPGASmartNICRDMAcollectivesASTRA-sim
Apr 2022 – Aug 2023
Undergraduate Researcher
CAAD Lab, Boston University, with Red Hat · Boston, MA
  • Machine-learning-guided LLVM optimization for FPGA high-level synthesis, funded by the Red Hat Collaboratory; published at LLVM-HPC 2022.

Education

Sep 2023 – May 2027 (expected)
PhD in Computer Engineering
Boston University · Boston, MA
  • Advisor: Prof. Martin Herbordt, CAAD Lab.
  • Research focus: LLM training and inference systems, HPC, distributed systems, networking, SmartNICs, and programmable switches.
En route to the PhD
MS in Computer Engineering
Boston University · Boston, MA
May 2023
BS in Computer Engineering, Magna Cum Laude
Boston University · Boston, MA
  • Minor in Computer Science; concentration in Machine Learning.

Teaching

Teaching assistant, Department of Electrical and Computer Engineering, Boston University, 2019–2026: ten semesters in total.

Course Title Semesters
EC311 Digital Logic Design 2
EC413 Computer Organization 4
EC440 Operating Systems 1
EC528 Cloud Computing 1
EK103 Linear Algebra 1
EK307 Electrical Circuits 1

Service

Journal reviewer: IEEE Transactions on Parallel and Distributed Systems (TPDS) · Future Generation Computer Systems (FGCS) · ACM Transactions on Reconfigurable Technology and Systems (TRETS)

Conference and workshop reviewer: IEEE FCCM · FPT · H2RC · IEEE HPEC · HEART · MIT URTC