Decagon · Enterprise platform for customer service agents
Product and access
Decagon combines cross-channel customer service agents with AOPs for business logic. Duet supports creation and optimization; technical teams control integrations, versions and guardrails.
Provider
Decagon develops AI agents for enterprise customer experiences. Jesse Zhang and Ashwin Sreenivas are listed as founders.
- Operator
- Decagon
- Organization
- Company
- Website
- decagon.ai
Core features
6 features from provider sources · checked on 10 October 2026. Availability and limitations follow below.
Describe workflows with AOPs
Teams define agent logic in natural language; AOPs connect those instructions with executable code.
Source 2Connect chat, voice and email
The platform brings multiple customer channels together in a shared agent logic.
Source 1Integrate support systems and knowledge
Connections provide customer data, knowledge articles, actions and escalations; APIs and MCP extend pre-built integrations.
Source 3Use Duet for agent development
Duet generates AOPs from past conversations and identifies gaps in workflows and guardrails.
Source 4Test and compare versions
Simulations, unit tests, versioning and A/B experiments support change validation.
Source 1Trace decisions and quality
The platform shows model calls, workflow triggers and referenced knowledge articles; Watchtower monitors conversation quality.
Source 1
Requirements and limits
Requirement
Prepare technical integrations
Customer information and actions depend on configured connections to CRM, helpdesk, telephony and knowledge systems.
Source 3Requirement
Define rules and escalations
AOPs need suitable rules for brand voice, escalations and agent behavior; technical teams manage integrations and versions.
Source 2Requirement
Review Autopilot updates
Duet Autopilot stages every update for human review with a health report.
Source 4Not yet confirmed
Clarify contract terms individually
The provider describes custom pricing. Unit prices, minimum volumes, implementation costs and contract duration require a quote.
Source 6
Operation & oversight
Operational details for this profile have not yet been researched.
Practical tasks
Editorial examples based on documented features. No product tests performed.
Handle order status with an AOP
- Your input
- Order system, support channel and rules for identity verification and handoffs.
- Possible result
- A customer service agent that retrieves order data and performs defined follow-up actions.
- What to check
- Validate data access, tool calls, edge cases and handoffs with test conversations.
Improve a support workflow with Duet
- Your input
- Past support conversations, existing AOPs and measurable quality goals.
- Possible result
- A proposed AOP update with tests and a review report.
- What to check
- Review proposed rules, simulations and impact before rollout.
Order system, support channel and rules for identity verification and handoffs.
Your possible prompt
Design a process for new requests: capture information, sort it by topic, and create a proposed response. Plan an approval step before external actions.
Getting-started idea based on documented capabilities. Not an executed product test. Adapt the task, data, and permissions.
Setup and pricing
Setup
Request a demo and quote, configure knowledge sources and integrations, and define and test AOPs
Prices and included usage
Pricing checked
Quote on request; the provider describes per-conversation or per-resolution billing
- Custom enterprise quote
- Custom quote
The provider’s own overview describes per-conversation or per-resolution billing and directs pricing enquiries to sales.
Price source
- The reviewed provider source publishes no fixed unit prices or allowances. Clarify billing definitions, minimum volumes and setup in the quote.
Sources and evidence
Official product documentation reviewed. Workflows and assessments are editorial; hands-on product tests are outside the scope of this directory.
Editorial team: Michael Meierhoff · sources reviewed on . Provider information is not independent evidence of performance.
-
decagon.ai /product/overview
-
decagon.ai /product/aop
-
decagon.ai /product/integrations
-
decagon.ai /product/duet
-
decagon.ai /about
-
decagon.ai /blog/ai-customer-service
