About

I am a Research Fellow at Microsoft Research India, working with the multilingual NLP group. My research focuses on cross-lingual parity in large language models, i.e. understanding why model behavior differs across languages, measuring those gaps, and developing methods to close them.

More broadly, I am interested in studying model behaviour, across languages and tasks, through the lenses of interpretability. I am especially interested in what mechanistic and behavioral tools reveal about why models perform unevenly, and how those insights should shape deployment of these models. I am interested in work where the aim is to translate findings from interpretability into concrete methods for closing these gaps. You can find my CV here

Publications

• DEPART: DEcomposing PARiTy across Multilingual LLMs. Findings of EMNLP 2026.
  A framework for decomposing the sources of cross-lingual performance gaps in multilingual LLMs. arXiv:2605.28163
• The State and Fate of Multilingual, Contextual Evaluation in the NLP World. published at COLM 2026.
  A data-driven survey of 51 multilingual benchmarks revealing that 36% of languages appear in just one benchmark, 56% of test data is translated from English, and contamination runs rampant in older suites. PDF
• The Role of Synthetic Data in Multilingual, Multicultural AI Systems: Lessons from Indic Languages. ACL 2026 (Main track).
  Developed UPDESH, a 9.5M-sample culturally grounded synthetic instruction dataset across 13 Indic languages to narrow performance gaps in multilingual AI. arXiv:2509.21294
• Garuda: An Indian Agricultural LLM and Krishi Saathi Advisory Chatbot. Workshop on Open-Source AI for Mainstream Use, AAAI 2025.
  Accepted as a hands-on demonstration paper.

Experience

• Research Fellow, Microsoft Research India — Multilingual NLP group. Cross-lingual parity, interpretability, and AI safety in large language models.
• Associate ML Scientist 1, Wadhwani AI — LLMs for Agriculture, domain adaptation, evaluation, agentic systems.

Education

• B.E. Electrical and Electronics Engineering with a minor in Data Science, BITS Pilani K.K. Birla Goa campus, 2020-2024

Get in touch

The best way to reach me is by email. Code is on GitHub; papers on Google Scholar.