For founders and engineering leads, daily standups can feel like a necessary evil: low-bandwidth status reports that eat up precious time and energy. But what if you could eliminate the prep work entirely and transform those meetings into high-value problem-solving sessions? Notion's Ryan Nystrom, an engineering manager, explains how their internal "Hot Potato" AI agent does just that.
This isn't about better note-taking. It's about a custom AI system that synthesizes team activity across Slack, tasks, and pull requests, then delivers a concise pre-read before the meeting. The goal is to move beyond the rote "I did this thing" updates and instead talk about problems, decisions, and wins. Claire Vo notes, "If you can have high bandwidth, high quality meetings with high frequency without the overhead, I think you can get better, more detailed work done collaboratively, especially if you're running like a remote team."
Nystrom shares how this changed his own workflow: “I can basically work up until like the minute of our meeting without having done a bunch of like prep. And then we all get on a video call and we look at this screen and we're like, 'Okay, here's what we need to talk about.'” The result? More relaxing, more fun, and more productive meetings. He calls it a "win-win-win."
Key Takeaways
- Automate Low-Value Prep: Notion's "Hot Potato" AI agent autonomously gathers updates from Slack, tasks, and pull requests, generating a meeting pre-read daily at 9:00 a.m.
- Shift Meeting Focus: By eliminating rote status updates, teams can dedicate standup time to discussing problems, making decisions, and celebrating wins, as Nystrom describes.
- Reduce Managerial Toil: This system dramatically cuts down on the time managers spend synthesizing information, freeing them to work right up until the meeting starts, while preventing burnout.
- Increase Collaboration Quality: Claire Vo highlights that high-bandwidth, frequent meetings, made possible by AI automation, drive more detailed collaborative work, especially for remote teams.
- Build Your Own Agent: The Notion AI Hot Potato Custom Agent Setup for Standup Prep provides a concrete framework for building a similar system.
The Notion AI Hot Potato Custom Agent Setup for Standup Prep
Here's how Notion configured its internal AI agent to automate daily standup preparation:
- Run Trigger: I have this set up to run at 9:00 a.m. every single day.
- Context Gathering Window: I only want you to look back for the last like 24 hours of activity.
- Utilize Sub-Agents: I'm explicitly telling it to use sub agents, which is kind of a sleeper feature in Notion AI.
- Map-Reduce Tasks: I ask it to kind of like fan out and do like a map reduce where I'm saying go use the honeycomb MCP to figure out what the latest metric is. look in our project channel and like find updates, feedback, questions. I tell it where the task database is and how to look for tasks within this project and then how to find yesterday's meeting.
- Output Format Template: I give it a template in the instructions where I'm like this is your format. I care about CI speed, decisions, progress, changes, bugs, questions, risks, a little bit of guidance on writing.
- Post to Slack: when it's done, I have it post to Slack and I like emphasize this like b I want it to be brief and fun and sometimes it's really corny and then sometimes it's like really good, and it'll be very quirky and just post this like link in our Slack channel and it's like, hey, here's your your pre-eread, some little quibble about like whatever.
- Access Control (View Only): I give it access to all these things. And I'm like, you can only view all of this stuff because I don't want it going and like modifying our task database, our project database.
- Access Control (Edit Specific Database): But this meetings database in particular, I'm like, you know, you can edit content because this is the one you're going to like write and update the page.
- Integrate with Metrics (MCP): I know exactly where this metric is. It's in honeycomb. And so I like just configured the MCP in notion. By the way, I like I like used the agent to like set itself up. I was like, here's the query. I literally gave it a screenshot of the honeycomb query. I was like, I don't know how this works. Can you just like update your instructions?
When This Works (and When It Doesn't)
This system works best when it mimics the ideal human process: if you had infinite time, every morning at 9:00 a.m., you'd comb through Slack, Honeycomb, GitHub, and task databases, asking questions and compiling a fun, brief summary. The AI agent simply scales that precise, thorough prep. It shines in teams with well-defined data sources (Slack channels, task trackers, metrics dashboards) and a consistent meeting rhythm. This method assumes you're already doing daily standups and have a history of relevant data to draw from.
It breaks down if your team's communication is scattered across too many unlinked tools, or if data sources are inconsistent. It also won't replace the need for managers to understand subtle team dynamics that aren't captured in digital records. If your team's "standup" is more of a casual check-in than a structured update, the overhead of setting up and refining such an agent might outweigh the benefits.
What to Do With This
This week, identify one recurring team meeting where prep feels like a chore. Say it's your weekly product sync, which requires collating updates from Jira, Slack, and your analytics dashboard. Using a tool like Notion AI (or a similar custom agent platform), map out your own version of the "Hot Potato" agent. Set the Run Trigger for an hour before the meeting. Define a Context Gathering Window that covers the last week. Give it Access Control to view your Jira board, relevant Slack channels, and a specific dashboard. Provide an Output Format Template that asks for key metrics, open blockers, decisions made, and upcoming priorities. Aim to automate 70% of the information gathering, freeing your team to discuss the implications of the updates, not just the updates themselves.