Rahil Parikh– Applied Scientist, AWS AI, Amazon

Rahil is an Applied Scientist at Amazon, currently working with AWS AI on developing and training AI agents, with a focus on enterprise workflow automation for Amazon Q. His work focuses on training, alignment, and evaluation of Large Language Models (LLMs).

At Alexa AI, he has led initiatives in Responsible AI, developing privacy-preserving ML models, de-biasing LLMs, and developing the first Model Inversion Attack for NLP Tasks. His expertise spans across the responsible development and deployment of AI systems, with a particular emphasis on making AI both powerful and ethical. While working for Alexa AI’s Edge-ML team, he worked on representation and multi-task learning for Acoustic Event Detection and Acoustic Scene Classification.

Beyond his core research, Rahil actively contributes to the broader AI research community as a reviewer for leading conferences including NeurIPS, ACL, and EMNLP. He has presented his work on AI safety and LLM alignment at various technical forums, sharing insights on responsible AI development through technical talks at industry conferences and academic institutions.

Prior to joining Amazon, Rahil worked with Prof. Shihab Shamma at the Neural Systems Lab and Prof. Carol Espy-Wilson at the Speech Communication Lab, University of Maryland. He worked on a multitude of speech and language processing tasks such as speech separation, universal source separation, speech production and speaker verification from 2018 and completed his graduate studies in 2022.

Prior to joining the in 2017, Rahil completed his undergraduate studies at Birla Institute of Technology and Science in 2017.