About

I am a graduate researcher at the Stanford IRIS Lab within the Stanford AI Lab (SAIL), advised by Chelsea Finn and Anikait Singh. I also work closely with Jack (Hao) Bai.

My research focuses on interactive decision-making and reasoning in autonomous agents. I am particularly interested in how LLM/VLM agents can self-improve and generalize through large-scale reinforcement learning. Specifically, I am investigating continual learning for web agents, aiming to build systems that can adapt and evolve within dynamic web environments over time.

Previously, I contributed to the development of RAPTOR, a state-of-the-art RAG system accepted to ICLR 2024.

Publications

Parth Sarthi, Salman Abdullah, Aditi Tuli, Shubh Khanna, Anna Goldie, Christopher D. Manning
ICLR 2024

Teaching

Graduate Course Assistant Fall 2025, Winter 2026
CS 229 Machine Learning
Graduate Course Assistant Spring 2025
CS 224S Spoken Language Processing
Graduate Course Assistant Winter 2025
CS 107 Computer Organization & Systems

Coursework

Artificial Intelligence

  • CS 224R: Deep Reinforcement Learning
  • CS 224N: NLP with Deep Learning
  • CS 231n: Deep Learning for Computer Vision
  • CS 229S: Systems for Machine Learning

Systems & Theory

  • CS 149: Parallel Computing
  • CS 161: Algorithms
  • CS 111: Operating Systems