Back to projectPublic page

Memory & Context: Long-Term Persistence

Explore how agents remember across sessions using vector databases and Firestore

Part of AI Agents: From Chatbots To Doers

4 blocks0 nested pages
Sign in to cloneBrowse project
Last updated Oct 29, 2025. Clone to remix or explore the blocks below.
ed232e48...
93a861b4...
content

Agent Memory: Beyond Conversation

Educational content slides

Agent Memory: Beyond Conversation

Chatbot memory (short-term): Conversation history within one session Close tab = forgot everything

Agent memory (long-term):

  • User preferences and context
  • Past tasks and outcomes
  • Knowledge gathered over time
  • Relationships between information Close tab = remembers you tomorrow

Example: Day 1: "I'm studying calculus for the AP exam" Day 7: "Find me practice problems" Agent: "Here are calculus practice problems focused on AP exam topics" ✓ Remembered: subject, purpose, difficulty level

54cfa73b...
quiz

Quiz: 3 Questions

Test your understanding with this quiz.

0 / 3

Why do agents need long-term memory beyond conversation history?

6be665e3...
feedback

Design Agent Memory

Complete this exercise and get AI-powered feedback.

Design Agent Memory

Create a memory system that balances utility (remembering useful context) with privacy (not storing sensitive data unnecessarily).

Consider: What makes interactions more helpful? What could be harmful to remember?

Progress5 of 7
PreviousMulti-Step Planning: Breaking Down GoalsNextBuild A Research Agent