← All drillsBuild Stateful Agents with LangGraph
What you'll be able to do- ✓Build a LangGraph graph from state, nodes and edges, and use reducers and context correctly
- ✓Give a chatbot memory with a checkpointer and threads, and show what it does without one
- ✓Give an agent a tool and route between the model and the tool with a conditional edge
- ✓Test an agent with a grader written in code, compare fixes by their effect, and set a limit on how long it may loop
- ✓Apply LangGraph to an industry problem: keep policy in code and language in the model, and test the policy on its own
- ✓Pause a graph for a person's approval with interrupt and resume, and test that nothing risky happens without it
Design agents that plan, call tools, and remember context, and check that they actually do — how real autonomous assistants are built.
⌁ Used by Klarna, Uber, J.P. Morgan, Replit, and LinkedIn for multi-agent / tool-using orchestration.
Start this internshipCreate an account to unlock the 7 sections, the workbench, and askThili.
BeginSections
1. Build Stateful Agents with LangGraph
🔒 locked2. Your first graph: state, nodes and edges
🔒 locked3. A chatbot that remembers
🔒 locked4. Tools and conditional edges
🔒 locked5. Does the agent use its tool correctly?
🔒 locked6. An industry problem: a refund desk
🔒 locked7. Pausing for a person
🔒 lockedDig deeper
📄ReAct: Synergizing Reasoning and Acting in Language Models (Yao et al., 2022)
paper🔗LangGraph overview — official docs
article🔗Groq documentation (the API used in this drill)
docsPart of these learning paths
I want to build my first AI agent
View path →