No Priors• Mar 12, 2026• 29mInterview

From Note-Taking App to AI Workspace: The Simon Last Interview

From No Priors

Simon Last•Co-founder, Notion

Executive Summary

Notion has fundamentally shifted its mission from being a tool for humans to perform work, to a platform for humans to manage AI agents that do the work.
The company pursues a model-agnostic strategy, positioning itself as the "Switzerland for models" to provide customers with access to the best frontier, open-source, and low-cost models (including several from China).
Notion has evolved its AI capabilities from a simple writing assistant (Feb 2023) to a cross-platform Q&A feature (Oct 2023) and now to a full agentic product with customizable, autonomous agents.
The adoption of AI agents is radically changing internal engineering practices, with the co-founder no longer writing code manually and the potential for engineers to be 100-1000x more productive.

9 quotes

Concerns Raised

The rapid pace of AI model evolution requires rewriting internal harnesses approximately every six months.
Building truly useful agents was a multi-year 'slog' that required several failed attempts before succeeding.
Agent-written pull requests are often larger and more complex, requiring robust testing and careful human review.

Opportunities Identified

Transforming Notion into the primary platform for managing AI agents in the workplace.
Achieving 100x-1000x productivity gains for engineers who effectively leverage coding agents.
Creating a unified semantic index across a company's disparate knowledge sources like Slack and Google Drive.
Enabling agents to bootstrap their own capabilities, such as building and deploying new software integrations on the fly.

Key Themes

The Agentic Shift

Notion's core vision has evolved from being a tool for human productivity to a collaborative workspace for humans and AI agents. The goal is no longer for users to do the work themselves, but to manage a swarm of agents that perform tasks on their behalf, using the same primitives like documents and databases for coordination.

This represents a fundamental paradigm shift in the purpose of productivity software, moving from direct manipulation to delegation and management, which has profound implications for the future of knowledge work.

Model-Agnostic Platform Strategy

Notion deliberately avoids dependency on a single AI lab, instead integrating a variety of models from different providers, including frontier, open-source, and low-cost Chinese models. This "Switzerland for models" approach provides flexibility and allows them to offer the best-performing or most cost-effective option for any given task.

This strategy de-risks their AI roadmap from the rapid, unpredictable progress of specific model providers and offers customers choice, preventing vendor lock-in and optimizing for cost and performance.

Iterative AI Product Development

Notion's AI journey demonstrates a pragmatic, phased approach, starting with a simple, single-step "AI Writer" before tackling more complex challenges. They then built a sophisticated Q&A feature with a semantic index across Notion, Slack, and Google Drive, which laid the groundwork for their full agentic product launched in late 2023.

This showcases a successful go-to-market strategy for complex AI products: start with a high-value, achievable feature to gain user adoption and technical learnings, then build towards the more ambitious, long-term vision.

Adapting Infrastructure for AI

To make their platform usable by AI, Notion had to treat the agent as a new type of customer and build infrastructure specifically for it. This included creating a new markdown dialect and an API allowing agents to interact with databases via SQLite, as well as rewriting their internal AI harness every six months to keep pace with model evolution.

This highlights that simply plugging an LLM into an existing product is insufficient; significant engineering effort is required to create agent-friendly APIs and data structures to unlock true agentic capabilities.

The Future of Engineering

The co-founder, Simon Last, has not manually written code since summer 2023, instead managing agents to perform development tasks. He believes AI tools can create a 100x or even 1000x productivity gap between engineers who leverage them effectively and those who don't, shifting the role from coder to 'agent manager' and verifier.

This provides a real-world example of how AI is transforming the software development lifecycle, increasing the scope and ambition of projects while fundamentally changing the day-to-day work of engineers.

Get started free

Topics

AI Agents Notion Productivity Software Knowledge Management LLMs Model Agnostic Engineering Productivity Software Development AGI Custom Agents Semantic Search RAG API Design Product Strategy AI Integration

Processed Mar 31, 2026 yt-dlp + mlx-whisper + Gemini