I Ran 157 Agent Plans Against a Real LLM. The Problem Wasn't Execution. It Was Planning.

8 relevance

Empirical study of LLM planning failures, directly addresses agent orchestration challenges.

AI/ML dev.to

Answer the board on AI readiness

Arrive with evidence, not opinion. See how your foundation benchmarks against your peers, get a clear read, and leave with a ranked place to start.

Why It Matters

For engineers building agentic systems, this shifts the optimization priority from tool-calling and orchestration to plan quality — a layer most teams skip, leading to brittle agents that fail after mutating state.

Key Takeaways

Implement a separate critic LLM with deterministic pre-flight checks on plan structure before any tool execution, and treat escalation to humans as a feature, not a failure.

I Ran 157 Agent Plans Against a Real LLM. The Problem Wasn't Execution. It Was Planning.

Summary

After running 157 agent plans against a real LLM, the core failure mode wasn't execution but planning — agents produce plausible-looking plans that miss critical dependencies or ordering constraints, leading to state corruption. The author built PlannerCritic, a system that treats plans like pull requests: a separate LLM critic reviews the planner's output, deterministic gates check ordering and rollback coverage, and the loop either converges on a safe plan or escalates to a human with a minimal question. This mirrors research showing self-correction fails when models can't independently verify their own output.

Author

Debashish Ghosal