
The Superintelligence Podcast
Notion’s AI agent workspace

Akshay Kothari
Co-founder, Notion
“Notion is not trying to build just for the Fortune 500. We really want to bring the Fortune five million forward.”
In this conversation
Notion co-founder Akshay Kothari joins Kim Isenberg and Peter Thum on The Superintelligence Podcast to discuss how people, code and AI agents could work together in the same workspace. They explore Notion’s template ecosystem, Notion Workers, internal agents such as “Smilers,” and the choice between different AI models. Akshay discusses automating routine work while retaining human judgment, maintaining useful knowledge bases and sharing specialized agents across an organization.
Key takeaways
Notion has rebuilt its agent harness roughly seven times in three and a half years, about every six months.
Notion Workers uses AI to write a small piece of code that syncs a Notion database with any outside data source about every 15 minutes, giving agents context without spending GPU time on each request.
A custom agent is defined by a trigger, its context and plain-language instructions. Notion’s internal people agent answers office questions in Slack and opens a ticket when it does not know the answer.
Open-weight models cost about one twentieth of frontier models, and Notion lets teams choose either per agent.
Software spend was about 2 percent of US GDP while knowledge work is about 40 percent, Kothari says. He argues software can now do the work, not just organize it.
More interviews
The daily AI briefing people actually read.
What shipped, who’s winning, and what it means. Six days a week, in under ten minutes.
Interested in a partnership?