Agent memory isn't one thing: sessions, wiki, procedural, and the forgotten 'judgment' layer
sujingshen · x · 2026-09-16
Responding to Pat Erichsen's article on the "memory trifecta" for personal agents, the author argues that saying "I added memory to my agent" is meaningless without specifying the layer:
- Session memory: what just happened in this conversation
- Wiki memory: long-term stable facts and knowledge
- Procedural memory: how work gets done — workflows/skills
Most products lump all three into one toggle. The author adds a fourth layer most people miss: judgment memory — why you rejected a proposal, what counts as "good enough." Systems often stuff this into chat summaries, and once summaries compress, the judgment is gone.
Self-check: run the same scenario (e.g., prepping a meeting) against all four layers; missing one means it isn't a real personal AI.
More from coding & agent
- Designer builds full accounting firm website end-to-end with Claude — Tegadesigns · 2026-09-16
- Unsloth launches desktop app to run and train LLMs locally — thisdudelikesAI · 2026-09-16
- 160 agents ran overnight: tokenmaxxing vindicated — kevinnbass · 2026-09-16
- Dev builds Chrome extension putting AI inside WhatsApp Web, seeks free beta testers — ManufacturerLast7833 · 2026-09-16
- How to Debug a RAG System: A Six-Step Framework from Retrieval to Guardrails — ashishllm · 2026-09-16
- Factory Life: Six free sessions on building with a team of AI agents, from Warp to Hermes — intellectronica · 2026-09-16