Save this video free

CMU AI Agents 2026: 4. Memory and Skills for Agents

Graham Neubig · 1:15:23 · Watch on YouTube

Part of Carnegie Mellon University 11-768. See all 14 lectures and search them

These notes were made with YouTube Collector. Paste any YouTube link and get the same: a summary, chapters and a searchable transcript.

Overview

Daniel Fried explains how agents can preserve useful experience beyond the context window by storing facts, trajectories, and reusable skills, then retrieving or revising that information for later tasks. He compares human-written and automatically induced skills, including Markdown-based Agent Skills, text workflows, and executable code, and surveys methods such as MemGPT, Mem0, Reflexion, Agent Workflow Memory, SkillWeaver, PolySkill, and ReasoningBank.

Key takeaways

Chapters

0:00 Why Agents Need Memory Across Tasks
4:40 Context Windows, External Stores, and Three Kinds of Experience
9:30 Human-Written Skills Versus Skills Learned from Experience
15:25 Agent Skills Files and Progressive Disclosure
24:00 SkillsBench Tests Whether Skills Actually Help
30:10 Creating and Improving Skills from Real Feedback
32:40 MemGPT and Mem0 for External Memory Management
39:00 Reflexion Feedback and Retrieval of Past Trajectories
44:10 From Reusable Subtasks to Code Skills
48:30 Online Learning and the Skill Lifecycle
51:30 Agent Workflow Memory Extracts Reusable Web Actions
57:00 Programmatic Skills: Testing, Reuse, and Efficiency
1:03:00 PolySkill Addresses Website-Specific Skill Failures
1:06:00 Learning from Failures, Judge Quality, and Memory Pruning
1:12:00 Training Skill Induction with Reinforcement Learning

Keep these chapters and the full searchable transcript in your own library.

Transcript excerpt

0:01 Okay, welcome back everyone. >> So, um the first assignment is out and uh as Graham said, we we have credits available for you to to work on it for those of you who are enrolled in the class. Do you folks have any logistic questions? Cool. Okay. So we talked in the last lecture about how agents can uh choose what to retain within their context window as a way of you know storing information within a task and using that to um to determine what's important to do better within the task. So a big part of today is going to be about how can agents go beyond the context window and store information across tasks in external stores um that will be useful for them to improve

0:50 performance on tasks in the future. And this is going to be skill induction. Um, and uh, we'll build on some some work that was done in the past on memory systems for agents. Um, and show how to apply those to to agents. But there's also another view on skills. And skills are a way for people to write instructions for an agent that can encode kind of like best practices or the knowledge that you have about how a task should be done. And these are actually going to be complimentary to each other. Um so we'll cover it just a small amount in this lecture but um you can imagine a system that um is sort of having a human in the loop um that's relying on the agent to uh propose some potential best practices from the

1:38 experiences that it has a person edit those and goes back and forth or vice versa like you know having a skill and instruction that's written by a person

This is the start of the transcript. Save this video free for the whole transcript, searchable and time-coded.

In this course

Related notes

Summary, takeaways, and chapters were generated by AI from the video's transcript and may contain errors. The video belongs to its creator, Graham Neubig.

Want the full transcript?

Save this video in YouTube Collector to get its complete searchable transcript, your own AI summaries, and a library that keeps every video you collect in one place.