I am a Research Scientist at Salesforce AI Research in Palo Alto, where I work with Dr. Shafiq Joty. Language models now produce fluent answers faster than anyone can check them, so most of my work is on verification: judging a model’s reasoning step by step rather than taking its conclusions on trust.
Hard2Verify, our ACL 2026 paper, asks whether verifiers can find the faulty step in a solution to an open-ended mathematics problem instead of only grading its final answer. SFR-DeepResearch uses reinforcement learning to train a single agent that reasons and searches on its own, and a related project synthesizes training data for web agents of progressively increasing difficulty. A second thread makes mixture-of-experts models cheaper to train, most recently in Least-Loaded Expert Parallelism at ICML 2026.
I started out at BITS Pilani, Goa Campus, where I studied computer science and led the Society for Artificial Intelligence and Deep Learning. From 2021 I worked with Ameet Deshpande and Karthik Narasimhan at Princeton NLP on data augmentation for low-resource languages, and the next summer built a gesture detector for TV news at Red Hen Lab through Google Summer of Code. My undergraduate thesis, at Microsoft Research India with Dr. Sunayana Sitaram, measured what compressing a multilingual model costs in fairness; that work became an ACL 2023 paper.
At the University of Texas at Austin I completed an M.S. in Computer Science, advised by Dr. Greg Durrett and Dr. Ying Ding. I spent my first year in the TAUR Lab, repairing model-written programs; the summer of 2024 at Salesforce on SFR-RAG; and the rest in the AI Health Lab, where I built MedHallu, an EMNLP 2025 benchmark for detecting medical hallucinations. After graduating in 2025 I returned to Salesforce — and to the question of telling correct reasoning from convincing reasoning.
Education
- University of Texas at Austin M.S. in Computer Science · advised by Greg Durrett and Ying Ding 2023 – 2025
- BITS Pilani, Goa Campus B.E. in Computer Science, minor in Data Science · thesis at Microsoft Research India 2019 – 2023
News
- Aug 2026New preprint: Privileged Likelihood Is Not Automatically Value — token-credit checks for on-policy self-distillation.
- Jul 2026New preprint: Mixture-of-Parallelisms — a memory-efficient training stack for trillion-parameter MoE models.
- Jul 2026Hard2Verify appears at ACL 2026 as a main-conference long paper.
- May 2026Least-Loaded Expert Parallelism is accepted at ICML 2026.
- Oct 2025Two new preprints: Hard2Verify, a step-level verification benchmark for frontier math, and Synthesizing Agentic Data for Web Agents.
- Oct 2025Spoke at Dreamforce 2025: “How to Build Flexible Deep Research Agents” — recording on YouTube.
- Sep 2025New technical report: SFR-DeepResearch — reinforcement learning for autonomously reasoning single agents.
- Aug 2025MedHallu is accepted at EMNLP 2025 (main conference).
- Jun 2025New preprint: EgoVLM — policy optimization for egocentric video understanding.
- May 2025Joined Salesforce AI Research as a Research Scientist, and graduated with an M.S. in Computer Science from UT Austin.
- May 2025New preprint: Teaching with Lies — curriculum DPO on synthetic negatives for hallucination detection.
Older news
- Feb 2025New preprint: MedHallu — a comprehensive benchmark for detecting medical hallucinations in LLMs.
- Jan 2025FaithEval is accepted at ICLR 2025.
- Sep 2024Technical report: SFR-RAG — towards contextually faithful LLMs.
- Jun 2024Joined Salesforce AI Research in Palo Alto as a research intern.
- May 2024AdaPT is accepted at NAACL 2024 (Findings).
- Aug 2023Started my M.S. in Computer Science at UT Austin.
- May 2023Our study on model compression and fairness is accepted at ACL 2023.
- Jan 2023Selected for Google Research Week; started a project at APPCAIR Lab in collaboration with TCS Research.
- Jul 2022Joined Microsoft Research, Bengaluru as a research intern.
- May 2022CIAug is accepted at NAACL 2022 and DMix at ACL 2022.
- May 2022Selected as a Google Summer of Code contributor with Red Hen Lab.
Publications
Also on Google Scholar.
2026
2025
2024
2023
2022
2020
Experience
Salesforce AI Research, Palo Alto, CA
May 2025 – presentResearch Scientist
- Research on reasoning and reliability in LLMs: step-level verification of mathematical reasoning (Hard2Verify, ACL 2026), autonomous deep-research agents (SFR-DeepResearch), and agentic data synthesis for web agents.
- Work on efficient large-scale training and inference of mixture-of-experts models (Least-Loaded Expert Parallelism, ICML 2026; Mixture-of-Parallelisms).
- Presented “How to Build Flexible Deep Research Agents” at Dreamforce 2025, and named on U.S. Patent Application 19/043,190, “Systems and Methods for Building Artificial Intelligence Agents.”
AI Health Lab, University of Texas at Austin
Jun 2024 – May 2025Graduate Research Assistant
- Introduced MedHallu (EMNLP 2025), the first benchmark designed specifically for detecting medical hallucinations in LLMs, built on 10,000 curated question–answer pairs with stratified difficulty levels.
- Showed that hallucination detection remains difficult for semantically subtle cases, and that domain knowledge and an explicit “not sure” option substantially improve detection.
- Developed Teaching with Lies, which trains hallucination detectors on synthetic negative examples using curriculum direct preference optimization.
Salesforce AI Research, Palo Alto, CA
Jun 2024 – Aug 2024Research Intern
- Worked with Dr. Shafiq Joty on SFR-RAG, a 9B retrieval-augmented LLM that outperformed models 10× its size on RAG benchmarks at the time of release.
- Curated a faithfulness-focused evaluation set that tests whether a model stays grounded in the context it is given, later released as part of FaithEval (ICLR 2025).
- Developed a “thought / observation” strategy for SFR-RAG that significantly improved multi-hop question answering.
TAUR Lab, University of Texas at Austin
Jul 2023 – Jun 2024Graduate Research Assistant
- Worked with Dr. Greg Durrett on efficient debugging of LLM-generated programs using feedback and error traces.
- Co-developed CodeUpdateArena, a benchmark for knowledge editing in code LLMs under synthetic API updates.
- Found that fine-tuning a code model on example usages of an updated API improved its problem solving more than updating it from documentation alone.
Microsoft Research, Bengaluru, India
Jul 2022 – Jan 2023Research Intern
- Worked with Dr. Sunayana Sitaram on compressing large language models with adapters, balancing computational efficiency against bias.
- Evaluated the zero-shot performance of compressed massive multilingual models, and deployed compressed variants for downstream use.
- Studied the effect of compression on fairness with both intrinsic and extrinsic measures (ACL 2023).
Google Summer of Code, Red Hen Lab
Jun 2022 – Sep 2022Contributor
- Built an end-to-end multimodal vision transformer to detect hand gestures co-occurring with time expressions in TV news, supporting accessible captioning.
- Framed the task as classification of body-keypoint trajectories, as part of Red Hen Lab’s work on multimodal analysis of TV news.
- Wrote up the approach and results in a blog post.
Princeton NLP
Jun 2021 – May 2022Research Collaborator
- Worked with Ameet Deshpande and Karthik Narasimhan on interpolative data augmentation (MixUp) for low-resource multilingual NLP.
- Studied whether interpolating between training examples improves transformer models on languages with little labeled data.