A team is implementing a multi-step Azure multi-agent workflow that must checkpoint intermediate state, branch on tool outcomes, and resume after human approval without re-running completed nodes. They prefer a graph-based open-source orchestration model that natively models nodes, edges, and durable state. Which framework choice best matches these requirements?
Select an answer to reveal the explanation.
Short Explanation
You need graphs, conditional branching, checkpoints, and resume-after-HITL. LangGraph is purpose-built for stateful multi-agent graphs with those capabilities (and the exam explicitly lists LangGraph among multi-agent orchestration frameworks). One giant stateless prompt has no real checkpoints. Blind cron restarts re-do completed work. CSS animations are not orchestration. C is the fit.
Full Explanation
Correct answer: C. Implementing multi-agent orchestration frameworks includes Agent Framework, LangChain, and LangGraph. LangGraph’s graph model with stateful nodes, conditional edges, and checkpointing supports branching on tool outcomes and resuming after human approval without replaying finished steps.
A is incorrect because a single stateless concatenated prompt lacks durable intermediate state and controlled resume.
B is incorrect because unconditional full restarts waste work and do not model graph checkpoints.
D is incorrect because UI simulation is not backend multi-agent orchestration.
Exam focus: map durable multi-step agent workflows with HITL resume to graph-based stateful frameworks such as LangGraph.