Week 6: Agent Frameworks
LangGraph, CrewAI, framework comparison and selection
Week 6 of 12
Learning Objectives
- Build agents with LangGraph
- Compare framework trade-offs
- Select appropriate frameworks for tasks
Topics Covered
- LangGraph state machines
- CrewAI team agents
- Framework comparison
- Orchestration patterns
Resources
Jupyter Notebooks
Required Readings
| Paper | Authors | Year | Link |
|---|---|---|---|
| LangGraph Documentation | LangChain | 2024 | Link |
| TaskWeaver: A Code-First Agent Framework | Qiao et al. | 2024 | arXiv |
Reading Guide: Agent Frameworks
Comparison of AutoGen, LangGraph, and TaskWeaver frameworks
Primary Paper
AutoGen: Enabling Next-Gen LLM Applications
Wu, Q., Bansal, G., Zhang, J., et al. (2023)
arXiv arXiv
Wu, Q., Bansal, G., Zhang, J., et al. (2023)
arXiv arXiv
Secondary Papers
Exercise: Framework Implementation
Build agents using LangGraph and compare frameworks
Learning Objectives
- Apply: Build agents with LangGraph
- Analyze: Compare framework trade-offs
- Evaluate: Select appropriate frameworks
Tasks
| Task | Points | Description |
|---|---|---|
| LangGraph Agent | 40 | Build research assistant with LangGraph |
| AutoGen Agent | 30 | Implement same agent in AutoGen |
| Comparison Report | 30 | Compare implementation experiences |
Exercise
Compare agent frameworks by implementing the same task in:
- LangGraph
- CrewAI
- Plain Python
Evaluate ease of use, flexibility, and performance.
Discussion Questions
- What are the trade-offs between different frameworks?
- When should you use a framework vs. building from scratch?
- How do you choose the right framework for your use case?
Additional Resources
Discussion & Questions
Join the Conversation
Have questions about this week's material? Want to discuss concepts with fellow students?