back_office_ops · saas · workflow

How Airtable built its AI agents framework powering Omni and Field Agents

Airtable's original AI features were limited to simple generative use cases and could not reason through problems requiring dynamic decision-making, retrieve additional data beyond what was provided upfront, or support a conversational interface for user feedback and follow-up.

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 · User message triggers agent loop
A user message event is produced by the user sending a message into the interaction, triggering the decision engine to call the backing LLM.
Tools used
OpenAI · partnerAnthropic · partner
Outcome

Airtable built a custom asynchronous event-driven state machine agentic framework that powers Omni and Field Agents, enabling reasoning, planning, and multi-step orchestration, with context optimization achieving a 15–30% reduction in tokens, inference latency, and cost.

What failed first

The original AI capabilities — the AI field, AI in automations, AI-generated select options, and AI formula generation — were incapable of dynamic reasoning or retrieving additional data, limiting them to straightforward generative tasks.

Results
Cost replaced~15–30%
Running since2024
Source

https://medium.com/airtable-eng/how-we-built-ai-agents-at-airtable-70838d73cc43

How we source this →

Grounding & classification
Source type: technical build writeup
23 fields verified against source quotes, 1 dropped as unverifiable.
agentic workflowai agentcontent generationconversational aisummarizationknowledge basehuman review describedmetric backednamed customerproduction runtime claimedtools describedworkflow describedsoftwarecost reductiontime savedtechnical build writeupback office opsagentic task execution