Write, Change, Recall, Forget: MongoDB's Pete Johnson on How Retrieval Drives Agent Performance

Today my guest is Pete Johnson, a 30-year technology veteran now serving as field CTO of AI at MongoDB. We start with a brief history of database technology, going back to the 1970 introduction of SQL and unpacking how the relative scarcity of disk space at that time informed data normalization as a design principle.

It goes beyond the title with direct discussion of like, data, that's, including: Today my guest is Pete Johnson, a 30-year technology veteran now serving as field CTO of AI at MongoDB.

1. 01As Pete recounts, in a remarkably short period of time, we've gone from being effectively forced to implement RAG pipelines by the very limited context windows of GPT-4 class model
2. 02Usage itself is scaling and naively maxing out the context window costs multiple dollars each and every time
3. 03Pete's key point above all is that agent performance and especially cost-adjusted agent performance depends heavily on effective retrieval