QA Engineering

Part 3: The AI Bug Taxonomy — 7 Failure Modes of Generated Code

← Previous Chapter: Part 2: Context Engineering | Series Hub | Next Chapter: Part 4: Multi-Agent Review Pipelines → Answer-first: AI-generated code suffers from distinct failure modes rarely seen in human junior developers: hallucinated API parameters, silent exception swallows, and plausible-looking but non-existent package imports (Package Hallucination).

Part 5: Agent Evals: Trajectory Validation & Automated Benchmarking

← Previous Chapter: Part 4: AgentOps | Series Hub | Next Chapter: Part 6: Human-in-the-Loop Gateways → Answer-first: Traditional single-turn evaluation metrics (BLEU, ROUGE) are useless for multi-step agents. Production eval pipelines evaluate Trajectory Efficiency (minimum tool steps to completion), State Invariant Compliance, and Negative Constraint Enforcement.

Part 5: Autonomous Testing & QA Automation at Scale

← Previous Chapter: Part 4: Legacy Refactoring | Series Hub | Next Chapter: Part 5: Operating Models → Answer-first: AI agents excel at generating property-based test assertions and exploratory edge-case inputs that human developers overlook, increasing critical path test coverage to 95%+.