Ayush Nangia

I work on agent evaluation and post-training, with a focus on long-horizon tasks.

I'm a research intern at Microsoft Research India. My research has included collaborations with SPAR and Jinesis AI Lab on post-training and AI safety.

In the past, I have done research at Proximal, Lossfunk, Toronto Metropolitan University through Mitacs, and Wadhwani AI.

If you're working on agent evaluations or post-training, get in touch.

Google Scholar/X/GitHub/LinkedIn