Worked on creating SFR-RAG 9B, which is a novel LLM, that outperformed (at the time of release) models 10x its size, like Command R+ and GPT4o.

It included a novel "thought" "observation" strategy which gave significant performance boost in Multihop QA tasks

Worked with Dr. Sunayana Sitaram on evaluation of zero-shot performance of compressed massive multilingual models.

Also worked on the effect on fairness of compressed vs original models, in both intrinsic and extrinsic measures.

Contributor to Red Hen Lab, project based on multimodal analysis 

Topic - Classification of body keypoint trajectories of gesture co-occurring with time expressions

Find more about the project on - Blog

Working on Data-augmentation techniques in NLP with Ameet Deshpande and Karthik Narasimhan

Working on knowledge graph and information extraction on medical data under the supervision of Dr. Ashwin Srinivasan.

Part of the group involved in research revolving around Deep Learning and Finance where we worked with data set of text and audio format and designed a multi‐modal architecture

Conducted post-training analysis that resulted in the finding of insights on model performance.

A core member of the AI group of BITS-Goa, whose main aim is to spread knowledge and learning related to Machine Learning to students.

The group also conducts brainstorming sessions on various trending topics in ML

Reading sessions where members present Research papers to others in order to be updated with the latest development in ML

Also a part of Open Source projects created by the group. 

Mentored newly inducted first-year students to get accustomed to the campus environment.

Guided them with their academics in their first year 

Helped to facilitate placements of 1000+ students