Oracle's Richmond Alake: Agents Differ by Memory, Not Models — 32K Beats 200K Tokens
AI Engineer · youtube · 2026-11-01
AI Engineer interview with Oracle's Richmond Alake, an early voice on agent memory. His thesis: everyone has the same models; agents differentiate via an "intelligence plane" — what an agent knows, remembers, is allowed to see, and how it improves. Using WWII codebreakers deducing "AF = Midway," he shows equal information ≠ intelligence.
- Demo: an ontology + semantic layer agent hits 99% accuracy with 32K tokens vs 200K for a naive agent
- Covers Oracle's agent memory package, deep data security, private inference
- Live fine-tuning of small decision models for memory routing
- Advice: agent = model + harness; own your memory layer
More from coding & agent
- Vibe-coded order flow trading terminal runs entirely in your browser — mdancho84 · 2026-11-01
- Google tests two screening policies to cut noise from agentic automated bug repair — burkov · 2026-11-01
- Travel Agent MCP Server serves up-to-date visa and immigration data for trip planning — modelcontextprotocol · 2026-11-01
- Parse MCP connector reveals how any brand surfaces in ChatGPT and Google AI Overviews — modelcontextprotocol · 2026-11-01
- Qwen3.8-Flash-Next under fire: strong at coding, falls apart on follow-up questions in research use — rm-rf-rm · 2026-11-01
- Dev claims Claude Opus 5.5 does in 1 day what 28 days of Unlimited Sol can't — karmay007 · 2026-11-01