Interpretability and reasoning infra for foundation models.
Envariant is an AI interpretability and reasoning SDK enabling foundation model builders to analyze, steer, and control their model's behaviors.
Founder
Working on interpretability infrastructure for foundation models! My background is in AI and bioengineering research at places like Stanford, MIT, Inceptive, and NASA.
