中文翻译
摘要: Harmix was a profitable music startup, before AI tools tempted them to build something new
Madison McLauchlan
April 13, 2026
Share on LinkedIn
Share on X
Share on Reddit
Share on Blue...
正文
Harmix was a profitable music startup, before AI tools tempted them to build something new
Madison McLauchlan
April 13, 2026
Share on LinkedIn
Share on X
Share on Reddit
Share on BlueSky
Harmix co-founder and CEO Nazar Ponochevnyi.
Startup raises $1 million to solve software fragmentation for SMBs.
Toronto-based
Harmix Group
got its start building an algorithmic search engine for the music industry. Along the way, the team was enamoured by another technical problem: fixing data fragmentation for small businesses using AI.
Harmix was an early entrant to the field of multimodal AI, or applying machine learning principles to different media. Founded in 2018, the company patented technology to let users search for images, video, and music using natural-language descriptions.
“When you have a large library of sometimes millions of compositions, there is no way you would have the time to scroll and to listen to all the previews,” co-founder Nick Shcherban said in an interview. Harmix’s search engine allows users to find the clip they’re looking for just by describing its vibe or uploading a reference file.
“We started developing agents internally to help us, because we had a small team.”
Nazar Ponochevnyi
Harmix Group
That proved particularly useful for music and television producers. The Harmix API has licensed its tech to media brands such as Red Bull and Sky TV, the latter of which used it to put together the music for
Game of Thrones
spin-off
House of Dragons.
A recent partnership with music rights management system OpenPlay also put Harmix’s API in the hands of decision-makers at Disney and Warner Brothers.
Co-founder and CEO Nazar Ponochevnyi, who is originally from Ukraine and a graduate researcher at Toronto’s Vector Institute, claimed this part of the company is profitable. But the 20-person company recently found inspiration for a new business direction elsewhere. “Once we had so much traction with the music, image and video sear
采集时间: 2026-04-13 20:16:23
AI机器学习多模态创业公司人工智能