Skip to content
Back to full roadmap
topicadvanced

Reflexion / Self-Critique Loop

Agent critiques its own output → writes lessons to memory → doesn't repeat the same mistake.

3 hours2 resources1 prereqs

Reflexion (Shinn et al., 2023): ReAct + episodic self-reflection.

3 layers:

  1. Actor — attempt the task with ReAct
  2. Evaluator — score the result (LLM-as-judge or rule-based)
  3. Self-Reflection — on failure, ask "why did I fail?" and extract a lesson

The lesson is written to memory. The next attempt starts with past failures as prefix. Model reasons: "Last time I did X, didn't work, this time I'll try Y."

+20% over baseline on HotpotQA. Especially strong on coding tasks.

Prerequisites

Resources(2)

Related steps