Everyone Has the Same AI Models. Here's What Sets Agents Apart - Richmond Alake, Oracle
Source
youtube.com
Author
AI Engineer
Date
Why it matters
Argues a governed semantic layer and AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → memory can cut context windowThe maximum amount of text a model can consider at once — its working memory for the current conversation or task.Full definition → from 200K to 32K tokenThe chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.Full definition → while keeping accuracy, and covers memory security and fine-tuned routing models.
Terms in this piece · Glossary
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.
token — The chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.
videoMost of Your Docs Readers Are Now AI Agents - Hahnbee Lee, Mintlify & Lincoln Murr, Coinbase
videoIf Your AI Scores 100%, Your Eval Is Wrong - Ramakrishnan Lokanathan, eslams.com
videoYour AI Agent Wants Your Data. Here's How to Lock It Down - Jeff Smith, Oracle
videoWhy We Rebuilt Microsoft Office From Scratch for AI Agents - Yijia Liang, Compound AI
