I am working on video, but specifically trying to make it easier for anyone to “direct” their own content creatively without having to know all the technical skills for editing. Near to mid term is to try to use AI tech to enable this, though there’s actually a lot you can do even without that. My background is in developer tools (I was product director for CI/CD at GitLab) and so I am also trying to bring in the best from that experience around how to make hard stuff easy and difficult stuff at least possible.

I suspect there will be a lot of companies in the near future that make AI models with permissive licenses easier to use, along different niches (like I am doing for video). It is a really promising area and picking a niche helps with papering over the challenges with general models.