🤯

Anti-Patterns: Reasoning

Failures in how agents think — hallucinations that compound, sycophantic agreement that overrides truth, and evaluations that don't evaluate.

3 patterns

Agent hallucinates a fact or tool output, then reasons on the hallucination as truth, compounding errors through every subsequent step.

Agent abandons its correct reasoning when a user pushes back, agreeing with the user even when the user is wrong.

Using LLM-as-judge without proper calibration, or relying on self-evaluation that is systematically overconfident.