Capital de TokensHow One Model Navigates 90,000 Tables
Watch on YouTube Estimated duration: ~4 minutes Thirteen tools. Ninety thousand tables. Alex Xu says OpenAI’s data agent works across a data platform of 1.5 exabytes using a single model and a deliberately small set of tools. Those numbers make his architectural point. A team facing that much data might expect to add more models, route
Estimated duration: ~4 minutes Thirteen tools. Ninety thousand tables. Alex Xu says OpenAI’s data agent works across a data platform of 1.5 exabytes using a single model and a deliberately small set of tools. Those numbers make his architectural point. A team facing that much data might expect to add more models, routers to choose among them, and specialized retrieval systems. Xu says OpenAI took a simpler route. He describes the agent as “pretty vanilla” by design, and points instead to the context supplied to its model. Xu bases the account on a conversation with Emma Tang, OpenAI’s head of data platform engineering. The question worth following is this: if the agent has one model and thirteen tools, how does it find its way through ninety thousand tables? Xu’s answer, in the post we have, is six layers of context.
Full episode: https://youtu.be/FUt9mr5HVzU
AI Generated Content: Script, voices and images were produced with AI.
#Shorts