WEBSITE ↗ · PUBLICATIONS ↗ · LINKEDIN ↗ · LET’S TALK ↗
ML / NLP RESEARCHER · NEW DELHI, INDIA · OPEN TO RESEARCH ROLES
I study when language models can be trusted—especially when evidence conflicts, inputs change form, or evaluations miss something important. I build benchmarks and systems to test those limits.
Previously: Nous Research · DRDO · Complexity Science Hub Vienna / Co-founded ClerkTree
PYTHON · RUST · PYTORCH · NLP · RAG · MODEL EVALUATION
Interesting questions deserve working code.
Bring me yours → · ORCID




