Hi, I'm Hakan. Building tools to fix my own frustrations.
I realized I jumped straight into discussions without actually introducing myself here. I am Hakan Efe. I spend most of my time building projects, messing around with different operating systems, and trying to figure out how video content actually works.
I usually build things to solve my own problems. Right now, that problem is figuring out why people click away from YouTube videos. I got tired of staring at standard retention graphs that give no context, so I started building ViewPulse to get actual editing feedback.
I will be launching it here later this week.
I'm glad to finally join this community and meet other makers. I would love to hear what everyone else is working on right now. Drop your current projects below so I can check them out.
Replies
welcome Hakan. standard retention graphs tell you where people left, not why, so the "editing feedback" framing is the right move. is ViewPulse correlating drop-off with specific edit events like cuts and pacing changes, or is it more about audio/pacing patterns across the whole video?
@sabber_ahamed Thanks Sabber. It actually focuses heavily on specific timestamp events.
ViewPulse syncs the second-by-second retention data from the YouTube API directly with the video transcript. When it detects a sudden drop-off curve, the AI looks at the exact context of what is being said and shown at that specific second.
So, if a creator is rambling on a static shot, the AI catches that local pacing error. It then generates specific editing instructions for that exact moment, like telling the creator to add a B-roll cut, a speed ramp, or a visual hook to bridge the gap. It is essentially hunting for the exact visual or psychological trigger that made the viewer click away at that second.