AI For Everyday Work
An evening on building the AI-native workplace, presented by AI Builders and Qoder in Singapore, with live demos from Notion, monday.com, Slack, and Zoom. I represented Notion alongside the Notion SG Community, sharing how Notion AI works best inside a team operating system, where documents, projects, and decisions stay connected instead of scattered across tools. Across the keynote, demos, and panel, the theme stayed consistent: the best AI setups reduce busywork, respect permissions, and still keep humans accountable for the final call.
What I covered
- How Notion started as a collaboration-first workspace, and why AI becomes more valuable when it supports team workflows, not just solo use
- The difference between on-demand agents and autonomous agents in Notion, from drafting and updating documents to ongoing QA, feedback triage, and task coordination
- Why the best agents are specific, scoped, and permission-aware, not generic do-everything assistants
- How multi-agent orchestration can work in practice, for example one agent summarising a meeting and another assigning follow-ups
- Why human review still matters, with AI handling repetitive work while people stay responsible for judgement and final decisions
- A live demo using Coda with Notion MCP to search a Notion workspace and turn one event page into an HTML presentation in one shot
Key takeaways
- AI is most useful when it is embedded where work already happens, not added as a separate layer
- Context, permissions, and business logic matter more than building a custom model for most teams
- Start with one painful, repetitive workflow and automate that well before expanding further
- Give agents the least access they need, and design them with a clear end-to-end scope
- Strong fundamentals still matter, especially critical thinking, communication, and accountability
Outcome
- The session sparked strong interest in practical, team-based AI use cases rather than abstract AI talk
- Conversations afterwards centred on Notion agents, secure AI adoption, and how lean teams can start small without overcomplicating their stack
- It was a strong reminder that calmer systems win, especially when AI helps reduce admin, context-switching, and scattered work