Evaluating & Testing Agentic Systems | Agentic Engineering | Oh My Learning

01 Why Agent Evaluation Isn't Unit Testing: Non-Determinism and Trajectories 10 min read mid drafted 02 What Benchmarks Measure and Their Blind Spots 11 min read mid drafted 03 LLM-as-Judge: Design, Calibration, and Known Biases 12 min read senior drafted 04 Trajectory Evaluation: Scoring the Path, Not Just the Outcome 11 min read senior drafted 05 Offline vs Online Evaluation: What Runs Before Ship vs in Production 11 min read senior drafted 06 Regression Testing for Prompts and Agent Behavior 15 min read staff drafted 07 Multi-Judge and Debate-Based Evaluation 14 min read staff drafted