EngineersofAI LogoCoursesBlogResearch LabAI LettersThe LabCode BankInteractive 3DKodrEarnestJobs
Skip to main content
Interactive 3D/ReAct Agent Loop
Goal
What is the capital of the country that won the 2022 World Cup?
THOUGHT
ACTION
OBSERVATION
Press Next Step or Auto-Play to begin
Task
Total steps8
Completed0
Iterations0
ReAct (Reasoning + Acting) interleaves thinking and tool use. The agent reasons about what to do (Thought), takes an action (tool call), observes the result, and repeats. This is the foundation of all modern AI agents.

ReAct Agent Loop - Interactive Visualization

ReAct (Reasoning + Acting) is the fundamental pattern behind every AI agent. The model alternates between thinking (generating a Thought about what to do), acting (calling a tool), and observing (processing the tool's output) - repeating until it has an answer. This visualization steps through complete ReAct traces for three realistic tasks using search and calculator tools.

  • 3 complete tasks: factual lookup (World Cup), calculation (compound interest), synthesis (LLM risks)
  • Step through each Thought → Action → Observation cycle manually or on auto-play
  • See how tool results are injected back into context to inform the next thought
  • Understand why chain-of-thought reasoning improves tool use accuracy
  • Foundation for: LangChain agents, OpenAI function calling, Claude tool use, AutoGPT
  • See iteration count - complex tasks require multiple reasoning-action loops

Part of the EngineersOfAI Interactive 3D - free interactive visualizations covering every major concept in machine learning and AI engineering. Hover any element for a plain-English explanation. No code required.