Bio
CBAI Research Fellow: mitigating post-hoc rationalization in LLMs using mechanistic interpretability.
Research Engineer at Cartesia (novel SSM-transformer TTS, CUDA optimization, synthetic data).
Internships at NVIDIA and Microsoft (LLM finetuning, RAG); BS CS + Math, University of Maryland.