April 04, 2025 — Cortex AI Observability — Preview¶
Snowflake announces the preview of AI Observability in Snowflake Cortex. You can use it to evaluate and trace your generative AI applications. AI Observability provides tools to systematically measure application performance, debug execution traces, and optimize configurations for production deployments. Key features include:
Evaluations: Assess application performance using metrics such as accuracy, latency, usage, and cost.
Comparison: Compare multiple evaluations side-by-side to identify the best configurations.
Tracing: Debug application executions by tracing inputs, outputs, and intermediate steps.
AI Observability supports task types such as Retrieval Augmented Generation (RAG) and Summarization. It provides metrics that you can use to evaluate context relevance, groundedness, and factual correctness.
For more information, see AI Observability in Snowflake Cortex.