Frontier multimodal AI
MGT combines 25 models to read the image, sound and text of videos: a rare field for applied research.
Frontier multimodal AI, real production stakes, brands and media companies using it every day. All our open positions are published on Welcome to the Jungle.
MGT combines 25 models to read the image, sound and text of videos: a rare field for applied research.
Brands, agencies and media companies producing and publishing with Aive, at scale, every day.
Aim high, and deliver it.
Build what does not exist yet.
See where video is going, before others.
Always do your best.
Founders who have scaled video companies before, a tight-knit team, and one motto: “Always do your best.”
On Welcome to the Jungle, with the details of each role and of the hiring process.
Yes, from the Aive page on Welcome to the Jungle.