Hi, I'm Sashank!
I'm a scientist with 10+ years of experience studying and mitigating the failure modes of intelligent systems. Currently, I lead the research team at atla ↗, where we develop critic models to evaluate & improve agents using modern RL & post-training.
Previously, I was an ML scientist at limbic ↗ building clinical AI throughout the mental healthcare pathway & ensuring its safety and reliability. Before that, I studied continual learning failures in the Niv lab ↗ at Princeton.
I enjoy making music and writing poems ↗. email: firstname[dot]lastname[at]gmail.com
