Raghu Hemadri

ML Research @ NYU

prof_pic.jpeg

370 Jay St.

NYU, Tandon School of Engineering

Brooklyn, NY, 11201

I am Raghu Hemadri, a Machine Learning Researcher with a strong focus on Natural Language Processing (NLP) and Reinforcement Learning (RL). Currently, I am working under the supervision of Prof. Siddharth Garg at NYU, exploring LLMs for hardware design. My latest research involves real-time hardware code analysis, a collaboration with Synopsys. Currently, I am collaborating with Microsoft Research on domain-specific LLM fine-tuning and distillation.

Beyond my research in LLMs, I also work with Siddharth Nayak (PhD @ MIT) on developing human-interpretable multi-agent RL (MARL), leveraging multimodal AI for effective coordination and decision-making.

Previously, I worked as a Member of Technical Staff at Oracle, where I developed ML-based log anomaly detection, automated bug detection, and large-scale recommendation systems.

Apart from my research, I am a professional badminton player. Badminton has instilled in me discipline, strategic thinking, and resilience—qualities that also translate into my research mindset. Beyond badminton, I enjoy playing and watching cricket.

I am currently on the job market for 2025 summer internships.

Contact Me

I am always open to discussing LLMs, RL, and engineering/science-related ideas. If you have an exciting idea, feel free to drop me an email. Also, if you’re up for a game of badminton or cricket, I’m always in—whether it’s playing or watching a great match!

news

Feb 06, 2025 Selected as the recipient of the David Chang and Cecilia M. Chang Student Leadership Award
Aug 23, 2024 Started a new position as Graduate Research Assistant under Prof. Siddharth Garg
Aug 23, 2024 Joined New York University for MS in CE! :sparkles: :smile:
Mar 18, 2024 Filed first ever US Patent
Jul 07, 2021 Joined oracle as Member of Technical Staff

latest posts

selected publications

  1. Detecting anomalies in log messages using code-derived message patterns to guide structured message classification
    Raghu Vamshi Hemadri, Arjun Upadhyay, and  al.
    US Patent, Mar 2024
  2. Iteratively Reweighted Minimax-Concave Penalty Minimization for Accurate Low-rank Plus Sparse Matrix Decomposition
    Raghu Vamshi Hemadri, Praveen Kumar Pokala, and Chandra Sekhar Seelamantula
    IEEE Transactions on Pattern Analysis and Machine Intelligence, Mar 2022
  3. ChildrEN SafEty and Rescue (CENSER) System for Trafficked Children from Brothels in India
    Raghu Vamshi Hemadri, Amarjot Singh, and Ajeet Singh
    Proceedings of the AAAI Conference on Artificial Intelligence, Jun 2022