Research Engineer with 4+ years of experience in LLM inference optimization, speculative decoding, and KV cache efficiency. Led projects across inference, serving, and infrastructure, delivering...