About
I'm an AI engineer and researcher based in Karachi, Pakistan, working on low-resource language models — primarily Urdu, Sindhi, and Pakistani NLP. My focus is speech synthesis, automatic speech recognition, and large language models for languages that have been systematically underserved by the field.
I work at the intersection of audio ML, LLMs, and RAG systems. Currently building at Proxima AI and 2DamnWav. 33 models and 25 datasets published on HuggingFace.
Research & Writing
