I'm a Senior Research Scientist on NVIDIA’s Nemotron team, pushing LLMs beyond static chat into the era of true autonomy. Working at the sweet spot of post-training, alignment, and test-time scaling, I build models that don't just talk, but reason through complex problems and generate code independently.

Before NVIDIA, I developed code intelligence models for Amazon Q Developer at AWS AI Labs. I studied at UCLA in the UCLA NLP Group (w/ Kai-Wei Chang), the University of Virginia, and Bangladesh University of Engineering and Technology. Deeply committed to the research community, I regularly serve as a (Senior) Area Chair, reviewer, and organizer at premier NLP and AI venues.

News and Announcements

1. [10.2026] We released Nemotron 3.5 Super and Lightning.
2. [08.2026] Open-SWE-Traces accepted to EMNLP 2026 (Findings).
3. [07.2026] Two papers got accepted at COLM 2026.
4. [06.2026] We released Open-SWE-Traces to advance distillation of SWE agents.
5. [06.2026] We released Nemotron 3 Ultra for Agentic Reasoning.
6. [05.2026] We introduce SWE-Zero to SWE-Hero framework!
7. [04.2026] GenCluster accepted to ACL 2026 (main conference).
8. [03.2026] I will serve as an area chair for NeurIPS 2026 and COLM 2026.
9. [03.2026] I will serve as a senior area chair for EMNLP 2026.