Position Awareness

Keep'em's AI isn't just aware of your content — it knows where each viewer is in the video right now. This enables contextual interactions that make the experience feel attended rather than automated.

How It Works

The video player continuously tracks the viewer's current position and sends this context with every chat message. When a viewer asks a question, the AI receives both the question and the current timestamp.

This means the AI prioritizes transcript content around the viewer's current position. If someone at minute 8 asks "can you explain that last part?", the AI looks at the transcript from roughly minutes 7–8 and provides a relevant explanation.

Examples

"What did he just say about the API?" — The AI checks the transcript around the current position, finds the section about the API, and provides a clear summary.

"How does this connect to what was shown earlier?" — The AI can reference content from earlier in the video and explain the relationship to the current section.

"I don't understand this part" — Even without a specific question, the AI can identify what's being discussed at the current timestamp and offer a simplified explanation.

Why It Matters

Position awareness is what makes the chat feel like a conversation with someone who is watching alongside you. Without it, the AI would treat every question in isolation, forcing viewers to provide explicit context. With it, the AI already knows the context — just like a human would.