Скачать или смотреть From Scripts to Agents: The New Stack for Testing AI

In this episode of Testing Tech Horizon, we explore how AI testing is shifting from scripts to agents. We cover Google’s open-source Gemini CLI, New Relic’s observability platform for LLM apps, Microsoft’s robotics foundation model (Rho-alpha), and Promptfoo’s new framework for evaluating stateful AI agents.


Research & References:

• Google Gemini CLI (Google Cloud Blog)
https://cloud.google.com/blog/topics/devel...

• Gemini CLI – GitHub Repository
https://github.com/google/gemini-cli

• Promptfoo: Evaluating Stateful LangGraph Agents
https://www.promptfoo.dev/docs/guides/eval...

• LangGraph Framework
https://github.com/langchain-ai/langgraph

• New Relic LLM Observability Platform
https://www.businesswire.com/news/home/202...

• Microsoft Research – Rho-alpha (Physical AI)
https://www.microsoft.com/en-us/research/s...

• NVIDIA Isaac Sim
https://developer.nvidia.com/isaac-sim

• LLM Observability Primer (Arize)
https://www.arize.com/blog-course/llm-obse...


Companion newsletter and episode notes:
https://daily.testingeducation.org/