Agentic AI · Systems Design

An agent that can't remember isn't autonomous. It's disposable.

The most overlooked problem in agentic AI isn't reasoning, tool access, or cost. It's memory — what was tried, what failed, and what comes next.

Memory console

0Lessons retained across tasks

The reset problem

Most agents today are goldfish. A task starts, the model gets a context window full of instructions, it acts, the task ends — and everything evaporates. Run the simulator with Persistent memory off: each task deposits lessons (the amber beads), and each task-end wipes them. The retained counter never grows. The agent re-discovers the same failures forever, and every run costs the same tokens to re-learn.

Flip persistence on and the beads survive into the episodic ring. By task three, the agent starts from accumulated experience — that difference, not smarter reasoning, is what turns a tool-caller into something you can leave alone for a week.

The four memory types (borrowed from cognitive science)

TypeWhat it holdsTypical implementationLifespan
WorkingThe current task: instructions, recent tool outputsThe context window itself (8k–1M tokens)One session
EpisodicWhat happened: past runs, failures, outcomesLogs + vector store, retrieved by similarityWeeks–forever
SemanticFacts about the world and the user ("deploys happen Fridays")Structured store / knowledge graph / notes fileLong-lived, edited
ProceduralHow to do things: learned skills, playbooks, prompts that workedSaved routines, fine-tunes, skill librariesLong-lived, versioned

Worked example: an agent asked to “fix the failing build” should consult episodic memory (“last Tuesday the same error came from a stale lockfile”), semantic memory (“this repo uses pnpm, not npm”), and procedural memory (“the lockfile-repair playbook”) — and only load the relevant slices into its limited working memory. Try shrinking the context-window slider: with fewer working slots, retrieval quality matters more, not less.

Enjoy this tool? Build your own with Super