5 quotes from 3 episodes on How I AI, each with a timestamped link to the source.
5 quotes3 episodes
The short version
Claire Vo states that companies succeed with AI agents by building project-level governance and automated loops. An autonomous agent will brute force a data warehouse with high-volume queries to find an answer.
Most interesting insights
Self-prompting workflows give AI agents their true value because manual human inputs defeat the purpose of automation.
“If your agent isn't able to prompt itself through an automation, what are you even doing?”
Claire Vo states that organizations turn chat folders into policy containers. Directly Responsible Individuals set custom rules for how teams use AI to finish specific jobs.
“What I haven't seen anybody talk about, which I actually think is really interesting, is using projects as a configuration layer and a governance layer on how your team actually uses AI to get a specific job done.”
Pointing an autonomous tool directly at a raw data warehouse leads to crashed systems. Claire Vo notes that an agent generates high-volume queries to find answers when unsure.
“Your data warehouse has to be very resilient to high volume queries because when in doubt an agent will just brute force it.”
Claire Vo describes a process where an automation sets a goal and runs the agent continuously. The system keeps working until it measures success or hits a block.
“…a type of loop that sets an outcome and runs an agent against that outcome until the outcome can be measured and validated or the agent is blocked.”
Notion’s internal “Boxy” system integrates an AI agent called Codeex to automate code generation and bug fixes directly from Notion tasks.
Engineers describe a task with natural language and a screenshot, @mention Codeex, and the AI produces a complete pull request with UI verification and a preview URL.
How we attribute quotes. Every quote was matched against the episode transcript, so the words and the timestamp are real (we trim filler words like "um", nothing else). The name comes from our written summary of the episode. YouTube gives us no voice-by-voice transcript, so open the timestamp to hear who is talking. See a wrong name? Tell us and we fix or remove it.