Reliable, self-improving enterprise AI
AI applications produce unreliable outputs and most models do not improve from their mistakes, causing errors in production.
Maitai provides an enterprise AI reliability platform that autocorrects faulty model output in real time and automatically fine-tunes models based on mistakes.
Developers and enterprises building and operating production AI applications who need more reliable inference and continuous model improvement.
Founder and CEO @ Maitai. Previous co-founder at Yew, led the AI engineering org at Presto (S10), and Big Bets team at Earnin. I spent most of my career getting AI applications from skunkworks to production, and then scaling them from there.





