Ph.D. student in Computer Science | Generative AI & LLMs researcher and engineer | builder of full-stack & agent systems
I study emergent behaviors in large language models and build end-to-end systems to experiment with, evaluate, and productize LLM research. I like shipping reproducible demos, creating clean READMEs, and making research easy to run and reproduce.
- Building Agent-Platform - an experimentation / orchestration platform for multi-agent research.
- Developing Council-of-LLM-s - experiments for multi-agent collaboration & theory-of-mind style research.
- Improving reproducibility for LLM experiments: training, evaluation, and deployment workflows.
(See my pinned repos below for details and quick demos.)
- Generative AI & Large Language Models (LLMs)
- Emergent behaviors, fine-tuning & adaptation, evaluation & robustness
- Multi-agent LLM systems and reproducible experimentation
Languages: Python, JavaScript/TypeScript, SQL Frameworks / Libraries: PyTorch, Hugging Face Transformers, React, Next.js, Node.js Infra: Docker, AWS, GitHub Actions (CI/CD), MongoDB Other: Model fine-tuning, prompt engineering, reproducible notebooks, system design
- Ph.D. in Computer Science, University of Louisiana at Lafayette (2025–Present)
- M.S., Computer & Information Systems, Rivier University (2023–2024)
- Portfolio: https://sanjayd.org
- Email: sanjayd3028@gmail.com
- Phone: +1 (475) 301 8229
- Open to: research-engineer roles, ML/LLM engineering, collaborations, open-source contributions, and guest talks.
- Best ways to reach me: email (above) or via my portfolio. If you want a quick intro or a demo, I’ll happily share a runnable Colab / Docker demo for any of the LLM projects.