Why it matters
NDC Copenhagen talk on evaluating, testing, and securing LLM applications, including RAG changes, prompt-injection resilience, harmful-response guardrails, Promptfoo, DeepEval, Vertex AI Evaluation, and LLM Guard.
My takeaway: Beyond the Prompt: Evaluating, Testing, and Securing LLM Applications - Mete Atamel is a model-evaluation signal. The practical read is to treat prompt and RAG changes as measurable system changes that need evals, guardrail tests, and regression coverage.