Kaaustaaub Shankar
I'm currently an MLE at Cybernetic Intelligence. Previously, I pursued my bachelors and masters at the University of Cincinnati under Dr.Kelly Cohen. I think we're heading towards a world where AI will be used in a lot of different places. The question is where should we use it and how can we make sure we use it safely. This is the main question that drives my work.
Always happy to hear reading suggestions or just talk — say hi.
interests
Interpretable AI
neuro-symbolic systems; models whose reasoning can actually be
inspected and audited.
AI control
Get as much capability out of a model in a safe manner
Capability evals
better ways to measure what LLMs can actually do, since control
depends on knowing that first.
reading