AI news · October 17, 2026
TwelveLabs launches Jockey for searching huge video libraries
Jockey ranked first on Product Hunt on October 16 and lets users query thousands of hours of video, returning cited moments and downloadable clip compilations.
- Product Hunt points when checked
- 372
- Product Hunt daily rank
- #1
TwelveLabs’ Jockey took the top spot on Product Hunt’s October 16 daily leaderboard with 372 points when checked. The product is built for media libraries: users can add video, ask for a specific event or object, and get an answer tied to clips with start and end times. The maker says it can search across thousands of hours rather than treating each file as a separate chat.
The launch page says Jockey can save an answer as an artifact or download an MP4 that stitches together cited clips. That is a useful output shape for editors, research teams, and social video work, but the launch score is only a traction signal. Search quality will depend on the event being concrete enough to find and on the user checking the cited moment.
What you can do with it
Try Jockey on a small, non-sensitive video set. Ask for five concrete events, open every cited timestamp, and record misses, false matches, query time, and the cost of the final clips. Keep a human check before publishing anything assembled by the tool.
Our take
Jockey is notable because it returns inspectable media evidence, not only a paragraph. That can shorten the path from search to an editor’s first cut. The commercial question is recall on the buyer’s own footage; a top launch rank cannot show that the citations are complete.
Links Try Jockey
Source: Product Hunt — Made With Models writes the brief; the reporting is theirs.