← All drillsBuild an Agent Memory Engine
What you'll be able to do- ✓Build long-term memory for AI agents
- ✓Extract and store durable facts from conversations
- ✓Resolve entities and consolidate contradictory memories
- ✓Retrieve memories by meaning, recency, and importance
- ✓Evaluate an agent's memory with a benchmark
Give AI agents persistent memory — the layer behind assistants that remember you across sessions, resolve who's who, surface the right fact at the right moment, and forget the noise.
⌁ Behind the "memory" features in ChatGPT and Claude; personalized assistants and tutors; customer-support bots that recall past tickets; and long-running autonomous agents — productized by memory stacks such as Maximem Synap, Mem0, and Zep.
Start this internshipCreate an account to unlock the 10 sections, the workbench, and AskThili.
BeginSections
1. Build an Agent Memory Engine
🔒 locked2. Why Agents Need Memory
🔒 locked3. Capturing & Extracting Memories
🔒 locked4. Entity Resolution
🔒 locked5. Storage & Embeddings
🔒 locked6. Retrieval Beyond Vector Search
🔒 locked7. Consolidation & Forgetting
🔒 locked8. Wire It Into an Agent
🔒 locked10. Encryption-First Storage (Optional / Advanced)
🔒 lockedDig deeper
📄MemGPT — Towards LLMs as Operating Systems (Packer et al., 2023)
paper📄LongMemEval — Benchmarking Chat Assistants on Long-Term Interactive Memory (Wu et al., 2024)
paper📄LoCoMo — Evaluating Very Long-Term Conversational Memory of LLM Agents (Maharana et al., 2024)
paper🔗thilimem — the reference implementation you build in this course
code🔗Mem0 — open-source memory layer for AI agents
codePart of these learning paths
I'm a software engineer and I want to learn agentic programming
View path →