# Shaurya Rohatgi - Personal Website

> A personal website showcasing Shaurya Rohatgi's profile, expertise, experience, education, projects, and publications. Built with Next.js, React, TypeScript, and Tailwind CSS. Focuses on LLMs, AI for Science, and Information Retrieval.

This site provides an overview of Shaurya Rohatgi's professional background and research interests. Key areas include:

- Large Language Models (LLMs)
- Retrieval-Augmented Generation (RAG)
- AI for Scientific Discovery
- Information Retrieval and Search Engines

The website is a single-page application built with modern web technologies.

## Key Information

- [Homepage](https://shaurya.ai/): The main content of the website. (Note: No `.md` version currently available)
- [Curriculum Vitae (CV)](https://shaurya.ai/assets/cv.pdf): Downloadable PDF of Shaurya's CV.
- [Google Scholar](https://scholar.google.com/citations?hl=en&user=UpHQFasAAAAJ&view_op=list_works&sortby=pubdate): Profile detailing publications and citations.
- [LinkedIn Profile](https://www.linkedin.com/in/shaurya-rohatgi): Professional profile on LinkedIn.
- [GitHub Profile](https://github.com/shauryr): Personal GitHub profile with projects.

## Optional

- [Website Source Code](https://github.com/shauryr/shaurya-bio-minimal): The GitHub repository for this website. (Note: URL based on local workspace name, may need updating if public repo differs)

## Experience

- **Research Scientist, Institute of Foundation Models, MBZUAI** (May 2025 - Present)
  - Building a gigantic high quality token collection which improve planning, reasoning, coding, and math abilities of the LLMs. 
  - Focusing on foundational research in large language models and their applications in scientific discovery and AI systems.
  - Responsible for experimentation and picking the best mixes of data for mid-training of LLMs.
- **Applied Scientist, AllSci** (2023 - May 2025)
  - Led the development of advanced LLM systems and RAG pipelines for scientific applications, with a focus on parameter-efficient fine-tuning and improving factual accuracy in domain-specific tasks.