What is Jockey by TwelveLabs?

Jockey is the first AI designed to understand your entire media library the way you do—searching by person, moment, or context across every photo and video you've captured. Powered by TwelveLabs' advanced model stack, it transforms scattered media into a single, instantly searchable archive. Whether you connect via MCP for Claude or ChatGPT, or build custom applications using the API, Jockey makes your media library accessible without manual tagging or folder digging.

Who it's for

• Marketing teams – who need to instantly find every ad, spot winning hooks, and track talent or logos across a growing creative library.
• AI-native creators – who want to talk to their footage in the assistant they already use, searching by plain language rather than filenames.
• Developers – who need a video understanding API to power search, extraction, and embedding features inside their own applications.

Key features

Natural-language search across your library

Describe a moment in plain language—"find all the hikes with mountain views"—and Jockey locates the exact clip, not the folder. It understands context, people, and scenes just as you remember them.

Connect to Claude (ChatGPT coming soon)

Plug your entire photo and video archive into the assistant you already use. Ask for a person, place, or moment directly inside Claude without learning a new interface. ChatGPT integration is on the way.

Multimodal embeddings and structured data extraction

Developers get one embedding across video, image, audio, and text to power custom retrieval and ranking. Define a JSON schema and receive timestamped, machine-readable metadata for every video—perfect for building your own search or analytics tools.

Automatic ad tagging and winning hook detection

For marketing teams, every ad is automatically labeled by hook, talent, format, and tone the moment it lands. Jockey also identifies which opening three seconds actually hold attention across your entire library.

What stands out

Jockey is the first AI that understands your media library just like you do—by person, moment, or context, not by filename or folder.

This isn't another tagging tool. Jockey actually comprehends the content inside your videos and photos, so you can search the way your brain works. It improves automatically with every update, and because it works through assistants you already use (Claude now, ChatGPT soon), there's no new workflow to adopt. For developers, the same intelligence is exposed as a full API, making video understanding embeddable anywhere.

Worth checking out if…

You have a growing library of photos, videos, or ad creatives that's becoming impossible to navigate manually. If you're a creator drowning in footage, a marketer who needs to surface the right ad in seconds, or a developer who wants to add video search to your product, Jockey offers a genuinely new approach—one that understands your media the way you do.