Ayush Rajesh Jhaveri
I am a second-year master’s student in Computer Science at NYU Courant Institute of Mathematical Sciences. My research interests lie in Natural Language Processing, and I will soon begin my thesis under the guidance of Prof. Eunsol Choi. Recently, I interned as an AI Engineer at BCG, and I previously worked as an SDE II at Amazon.
Research Projects
Studying whether pretraining perplexity can predict downstream performance in LLMs.
Identifying attention patterns under induced uncertainty and calibrating them in LLMs to reduce sycophancy.
Proposed a novel LLM evaluation framework (MAPLE score) to quantify math reasoning errors. arXiv link
Ayush Jhaveri, K. S. Sangwan, Vinod Maan, Dhiraj. Deep learning-based mosquito species detection using wingbeat frequencies. Intelligent Data Engineering and Analytics, Springer Nature, pp. 71–80. Publisher link
Education
- Courses: NLP, Emerging topics in NLP, Deep Decision Making & Reinforcement Learning
Experience
- Designed and built an agentic AI-powered price recommendation system using LLMs for a major telecom client.
- Led the design and development of a .NET-based vision product identification app for Amazon physical stores.
- Built AWS-powered device provisioning automations.
- Developed Vue.js-based payment systems.