Providing Greater Access to LLMs with Brandon Duderstadt, Co-Founder and CEO of Nomic AI
Gradient Dissent: Conversations on AI
00:00
Quantifiable Metrics for Quality Improvement
In the early iterations of these systems, they often make several quite obvious errors. So yeah, it's really setting up the data flywheel and getting that sort of like quality loop closed is what you'll want before. And in that process, how are you evaluating if things are getting better? feedback from users. Feedback from us internally, if we find useful feedback from the people that are using Atlas every day.
Play episode from 47:10
Transcript
