Design LLM workflows that plan, take actions, observe results, and loop — turning a chatbot into an autonomous problem-solver.
Five passes over the same idea, each from a different angle. Do them in order, or jump to whichever you need.
Agentic workflows move beyond single-shot prompt → response into a loop of reasoning, tool invocation, observation, and re-planning. The key abstraction is the "thought–action–observation" cycle popularised by ReAct. Building reliable agentic systems requires careful tool design, loop-break conditions, and idempotent side effects — or your agent will spend your budget spinning in a hallucination loop.
