Recursive Self-Improvement Meets a Strong Baseline Agent
Experiments with OpenAI agents examine correctness headroom, efficiency gains, and whether autonomous revision earns its added complexity.

Experiments with OpenAI agents examine correctness headroom, efficiency gains, and whether autonomous revision earns its added complexity.

Turn a favorable engineering idea into a clear decision: start, keep going, change direction, or stop.

How to decide when multiple agents justify their added time and cost.

Explore how GPT-6.1 Sol delegates to subagents through the Responses API and when the gains in accuracy and speed justify the token cost.

Using Codex and Promptfoo to define success, diagnose failures, and choose models

An evaluation guide to accuracy, latency, and cost across complex workloads.
