Ayush Agarwal

Ayush Agarwal is a research scientist in Applied AI Research at NVIDIA. His work focuses on agent optimization, model routing, inference systems, and diffusion models. Since joining NVIDIA in January 2025, he has contributed to NVIDIA NIM, Dynamo, agentic evaluation efforts, and model-routing research in NeMo Switchyard. Across these projects, he works at the intersection of AI research and the systems used to evaluate and serve models in practice. He received his master's degree in Machine Learning and Data Science from Northwestern University in December 2024.
Avatar photo

Posts by Ayush Agarwal

Decorative image.
Agentic AI / Generative AI

Route AI Agent Workloads Across Models with NVIDIA NeMo Switchyard

Building an AI agent does not end with choosing a single model. Each model has its own strengths, weaknesses, and cost profile, which can shift from one... 12 MIN READ