AutoContent API is an independent service. It is not affiliated with, sponsored by or endorsed by Google. Gemini, Gemini Notebook and NotebookLM are trademarks of Google LLC.

Docs to video: Kubernetes documentation as an explainer video

Docs to video with an API: the Kubernetes page on Pods became a narrated explainer for new developers. Watch it, copy the request, see its cost and our check.

How this was made. Audio and video overviews, slide decks and infographics are made with Google's Gemini Notebook (formerly NotebookLM), on accounts that AutoContent operates. AutoContent has no agreement with Google. Google can limit or stop this access at any time.

AI-generated with AutoContent APIDownload MP4 Full transcript (TXT)

What came out, and what we would change

We asked for an explainer for developers who are new to Kubernetes: why Pods exist, what runs inside one, how Pods are created and replaced, and one common mistake to avoid. The result is a 1280 × 720 MP4 of 9 min 16 s with numbered sections, short captions, and the AutoContent badge in the corner.

The content follows the page: a Pod is the smallest deployable unit and models a logical host; most Pods run one container, some run several that work together; containers in a Pod share namespaces, volumes, one IP address and localhost; init containers and ephemeral containers; Pods are disposable and are managed by workload resources such as Deployments, StatefulSets, DaemonSets and Jobs through Pod templates. The mistake it warns about is the one the page names: creating bare Pods, which nothing recreates when their node fails.

Two points are too strong. The video says Pods are never updated in place. The page says a changed template creates new Pods, but also that some fields of a running Pod can be updated in place. The video also calls shared localhost "deeply secure", which the page does not say.

For a whole docs site, send one request per page from a script: each video stays focused on one page, and a webhook tells you when each one is ready. An explainer video costs 50 credits; this one took 17 min 32 s to generate. Set style to a preset such as whiteboard or watercolour, or describe your own style in up to 5,000 characters.

Transcript excerpt

Speech-to-text of the output, not edited and without speaker labels. It is part of the sample output.

directly onto a server. But you actually don't. In this ecosystem, Kubernetes completely abstracts the container away from you. You manage pods, not raw containers. The pod is your absolute lowest level building block, which literally means you cannot deploy anything to a cluster that is smaller than a pod. To really wrap your head around this abstraction, it helps to think of a pod as a logical host. It's basically a wrapper that groups and isolates resources. You know how a traditional virtual machine back in the day would host a few tightly related

The exact request

The body that was sent to POST /content/Create, unchanged, and the same request as curl.

request.json

{
  "resources": [
    {
      "type": "website",
      "content": "https://kubernetes.io/docs/concepts/workloads/pods/"
    }
  ],
  "outputType": "video",
  "format": "explainer",
  "text": "Explain what a Pod is to developers who are new to Kubernetes: why Pods exist, what runs inside one, how Pods are created and replaced, and one common mistake to avoid.",
  "language": "English"
}

curl

curl -X POST "https://api.autocontentapi.com/content/Create" \
  -H "Authorization: Bearer $AUTOCONTENT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"resources":[{"type":"website","content":"https://kubernetes.io/docs/concepts/workloads/pods/"}],"outputType":"video","format":"explainer","text":"Explain what a Pod is to developers who are new to Kubernetes: why Pods exist, what runs inside one, how Pods are created and replaced, and one common mistake to avoid.","language":"English"}'
# then poll until "status" is 100:
curl -H "Authorization: Bearer $AUTOCONTENT_API_KEY" "https://api.autocontentapi.com/content/Status/<request_id>"

How we checked it

Transcript: We transcribed the output with Whisper large-v3-turbo and compared the transcript with the source.
Language: Detected en, requested en.
Loudness: -19.8 LUFS integrated, true peak -0.9 dBTP.
Notes: The concepts and the warning match the page. "Never updated in place" and "deeply secure" overstate the page.
Checked on: 2026-11-02. Failed generations are run again and never published.

More examples