While working on another startup, the coordination meetings were killing us. The information coming out of them was disjointed, nothing was actionable, and the same topics resurfaced every week.
We realized group meetings create the same distortions every time:
• Loudest voices dominate the conversation.
• Team members often self-censor.
• There’s internal pressure to align.
• Critical blockers stay private because surfacing them feels uncomfortable.
The current batch of AI tools treat these issues as a transcription problem. But a bad meeting with lots of noise just produces a bad transcript with lots of noise, and there’s nothing AI can do about that.
So, we built an internal tool for ourselves to address the problem.
Instead of scheduling another group call, AI speaks with the organizer to understand the goal and direction for the meeting, then privately and asynchronously interviews each participant. It asks follow-up questions, identifies disagreements and blockers, and then synthesizes everything into a single output with decisions, areas of consensus, and risks.
A few things we didn’t expect while using this:
• Team members were far more honest in private AI conversations
• Blockers, risks, and dissent surfaced more readily that had never been raised in group calls.
• Many of our recurring internal meetings disappeared, either because they weren’t necessary or folks opted to use our async meeting tool because it was faster and easier.
We found the tool so helpful that my co-founder and I turned it into an iOS app called Noada, and it's free on the App Store today. We're launching publicly because we want to see if others find it as helpful as we did.
The thing we're genuinely curious about is whether this model works for decisions that require real-time debate or brainstorming. We’ve tried to account for all the various types of meetings during development, but we still think async private input is best for alignment, status, and discovery.
elliot952•2h ago
We realized group meetings create the same distortions every time: • Loudest voices dominate the conversation. • Team members often self-censor. • There’s internal pressure to align. • Critical blockers stay private because surfacing them feels uncomfortable.
The current batch of AI tools treat these issues as a transcription problem. But a bad meeting with lots of noise just produces a bad transcript with lots of noise, and there’s nothing AI can do about that.
So, we built an internal tool for ourselves to address the problem.
Instead of scheduling another group call, AI speaks with the organizer to understand the goal and direction for the meeting, then privately and asynchronously interviews each participant. It asks follow-up questions, identifies disagreements and blockers, and then synthesizes everything into a single output with decisions, areas of consensus, and risks.
A few things we didn’t expect while using this: • Team members were far more honest in private AI conversations • Blockers, risks, and dissent surfaced more readily that had never been raised in group calls. • Many of our recurring internal meetings disappeared, either because they weren’t necessary or folks opted to use our async meeting tool because it was faster and easier.
We found the tool so helpful that my co-founder and I turned it into an iOS app called Noada, and it's free on the App Store today. We're launching publicly because we want to see if others find it as helpful as we did.
The thing we're genuinely curious about is whether this model works for decisions that require real-time debate or brainstorming. We’ve tried to account for all the various types of meetings during development, but we still think async private input is best for alignment, status, and discovery.
Would love to hear your thoughts. Thank you!
Landing page: https://www.noada.app/ App Store page: https://apps.apple.com/us/app/noada/id6757136534