back_office_ops · saas · workflow
How Notion Built Custom Agents: Rebuilding Four or Five Times to Reach an Agent-Native System of Record
Notion's early attempts to build agents starting in late 2022 failed repeatedly because there was no tool-calling standard, context windows were too short, frontier models were not reliable enough, and too much complexity was exposed to the model.
How it works
Common implementation structure
How this type of workflow is generally built, generalized across documented cases — not tied to any one vendor's stack. Click any stage to read what happens there. Specific products that implement these stages appear in “Tools commonly seen” below.
Stage 1 · Email triage trigger
A Notion Custom Agent begins by triaging incoming emails for coworking space tenant applications.
Tools used
NotionMCPCLIGeminiGPT fourfireworks
Outcome
After being rebuilt four or five times, Notion Custom Agents launched as Notion's most successful product launch in terms of free trials and conversions.
What failed first
Before a function-calling standard existed, Notion tried to fine-tune frontier models with a custom tool-calling framework and pursued partnerships with multiple frontier labs, but neither approach produced robust production-ready behavior.
Grounding & classification
Source type: technical build writeup
28 fields verified against source quotes.
agentic workflowai agentknowledge searchmulti agent workflowsummarizationemailknowledge basemeeting recordingbuilder submittedfailure mode describednamed customerproduction runtime claimedtools describedworkflow describedsoftwareconversion increasetechnical build writeupback office opsagentic task executiondata sync enrichmentinbox to system