Lakshya A. Agrawal

Why Do Multi-Agent LLM Systems Fail?

arXiv 2025

Multi-module GRPO: Composing Policy Gradients and Prompt Optimization for Language Model Programs

arXiv 2025

GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning

arXiv 2025

Guiding Language Models of Code with Global Context using Monitors

arXiv 2023

Dan Klein

Matei Zaharia

founder

Christopher Potts

Dilara Soylu

grad-student

Ion Stoica

professor / co-founder

Meng Jiang

Noah Ziems

Omar Khattab

Aditya Kanade

Aditya Parameswaran