---
type: "raw_signal"
id: "how-an-ai-agent-automates-qa-for-the-f8f3e5"
slug: "how-an-ai-agent-automates-qa-for-the-f8f3e5"
title: "How an AI Agent Automates QA for the Cerebras Cloud Console"
description: "Cerebras says it built an internal QA and release pipeline for its Cloud Console in which plain-English `.pt` test files are executed by an LLM agent in a real authenticated browser via Playwright MCP."
retrieval_nugget: "Cerebras says it built an internal QA and release pipeline for its Cloud Console in which plain-English `.pt` test files are executed by an LLM agent in a real authenticated browser via Playwright MCP."
observed_at: "2026-09-23"
record_date: "2026-09-23"
date_kind: "observed_at"
why_it_matters: "For UI-drift failures, the agent may propose a repair, but Cerebras says deterministic code applies a literal find-and-replace, requires a live rerun, and leaves merge approval to a human. Its release-risk floor may be raised by the model but not lowered."
novelty: "notable"
verification_level: "source-linked"
signal_type: "research"
source_platform: "cerebras.ai"
topics: []
entities: []
related_patterns: []
source_url: "https://www.cerebras.ai/blog/we-taught-an-ai-agent-to-qa-our-cloud-console-in-plain-english"
source_urls: ["https://www.cerebras.ai/blog/we-taught-an-ai-agent-to-qa-our-cloud-console-in-plain-english"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "signal:how-an-ai-agent-automates-qa-for-the-f8f3e5"
status: "published"
visuals: []
routes: {"html":"https://newruntime.com/signals/how-an-ai-agent-automates-qa-for-the-f8f3e5/","markdown":"https://newruntime.com/signals/how-an-ai-agent-automates-qa-for-the-f8f3e5.md","json":"https://newruntime.com/signals/how-an-ai-agent-automates-qa-for-the-f8f3e5.json"}
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
---

# How an AI Agent Automates QA for the Cerebras Cloud Console

## Retrieval answer

Cerebras says it built an internal QA and release pipeline for its Cloud Console in which plain-English `.pt` test files are executed by an LLM agent in a real authenticated browser via Playwright MCP.

Cerebras says it built an internal QA and release pipeline for its Cloud Console in which plain-English `.pt` test files are executed by an LLM agent in a real authenticated browser via Playwright MCP. The system records runs, generates tests from QA plans and DOM baselines, analyzes PR coverage against critical user journeys, and routes nightly failures.

For UI-drift failures, the agent may propose a repair, but Cerebras says deterministic code applies a literal find-and-replace, requires a live rerun, and leaves merge approval to a human. Its release-risk floor may be raised by the model but not lowered.

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- collection: [Raw Signals](https://newruntime.com/signals.json): Open every source-linked raw signal.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)