Delight.ai vs Twilio (2026)

Delight.ai integrates with Zendesk, Salesforce and Freshworks for live agent handoff, with Salesforce also supported as a knowledge source and third-party knowledge base integration so the agent can surface the right information in real time. Agents take action as well as answer, and Trust OS logs every tool call. On the channel side the agent reaches customers through in-app chat, web, email, SMS, WhatsApp and social, all built on Sendbird's communications platform, and an AI Agent Integrations page carries the connector set.

SourceDelight.ai integrations page, AI agent platform page and Salesforce and Freshworks product releasesread 2026-09-05

Agent Connect gives an agent a tool system built on standard function calling. A developer defines a name, description and JSON schema, and the SDK formats the tool for whichever LLM provider runs the agent. Custom tools are written with a decorator in Python or defineTool in TypeScript and can read the session, the customer profile and the channel.

Three tools come built in, knowledge search over Enterprise Knowledge, memory retrieval and escalation to a Studio flow that can route into Flex. Because each tool is the customer's own code, the agent can act in any system the customer can reach, and Twilio's own messaging, voice, email and Verify APIs sit alongside as further actions.

Sourcetwilio.com/docs/conversations/agent-connect/core-conceptsread 2026-10-11

Agent Steward takes full ownership of complex, long-running customer issues end to end instead of answering and closing, and its agents understand intent, hold context, take action and adapt, resolving complex issues and escalating only when needed.

Behavior comes from a customer-authored actionbook holding rules such as refund thresholds and escalation logic, which Openbook renders as a visual flow through complex branching.

Above the individual case, Zero-Touch Improvement runs a closed loop in which observability catches an issue, control tests the fix and oversight holds it to policy, in a graduated system that moves from human-guided to autonomous.

SourceDelight.ai Agent Steward, Trust OS and Omnipresent pagesread 2026-09-05

Conversation Orchestrator joins voice calls, SMS and WhatsApp messages into one threaded conversation tied to a customer profile, with a lifecycle that moves each conversation from open to inactive to closed on timeouts the customer sets. Studio, Twilio's visual flow builder, runs multistep communication workflows and receives the agent's handoff payload, and Flex routes the conversation to a person with the history attached. The reasoning inside the agent stays in the customer's own code, while the movement of a conversation between channels, the agent, flows and people runs on Twilio.

Sourcetwilio.com/docs/airead 2026-10-11

Channels are in-app chat, website, email, SMS, WhatsApp, social messaging and voice, all carried on Sendbird's communications platform, with context following customers between channels instead of restarting. Omnipresent agents are always on, always aware and always personal, and they start conversations themselves, anticipating needs before an issue forms and re-engaging when a customer goes quiet or switches channel. Voice 2.0 adds multilingual real-time outbound calling.

SourceDelight.ai Omnipresent, Voice AI Agent and AI agent platform pagesread 2026-09-05

Inbound voice calls, SMS, WhatsApp, RCS and web chat all reach the agent through one shared callback, so a single handler serves every channel. Webhooks and WebSocket connections carry the events, and Conversation Relay streams a live phone call to the agent with speech recognition and synthesis handled by Twilio. Outbound sends and calls start from the REST APIs, and Studio flows start on incoming calls, messages or API requests, which covers inbound, event driven and programmatic starts across the channels a contact center uses.

Sourcetwilio.com/docs/conversations/agent-connect/core-conceptsread 2026-10-11

Delight.ai integrates third-party knowledge bases so the agent can find and surface the correct information in real time, with Salesforce supported as a knowledge source and Freshworks and Zendesk connected alongside. Building an agent works like onboarding a team member, transferring knowledge to it through the supervisor dashboard, and the actionbook holds the authored rules the agent answers within. Trust OS detects factual errors and off-policy responses automatically, and the review-and-grade surface scores knowledge accuracy.

SourceDelight.ai AI agent platform and builder pages, and the Salesforce knowledge-source releaseread 2026-09-05

Enterprise Knowledge stores a company's policies, documents and reference material as knowledge bases that agents and human agents search for grounded answers. It is billed as a persistent store, $0.018 per GB hour of storage and $0.005 per retrieval, and Agent Connect exposes it as a knowledge search tool the agent calls when a customer asks a question, returning matching passages for the reply. Customer profiles live separately in Conversation Memory.

Sourcetwilio.com/en-us/products/conversational-ai/pricingread 2026-10-11

The Agent Memory Platform, sold as its own product, is the intelligence layer that gives agents a living memory and a business brain, connecting what the agent remembers about each customer with what the business is trying to achieve.

Agents absorb signals from every conversation and action to form an evolving understanding of the customer that goes beyond static CRM data, and context is kept as a customer moves between chat, SMS, email and voice so nobody repeats themselves.

The agent remembers every interaction, and continuity carries across separate conversations as well as within one, unlike conventional chatbots that lose history the moment a session ends.

SourceDelight.ai Agent Memory Platform and AI agent platform pagesread 2026-09-05

Conversation Memory keeps one profile per customer inside a memory store. Each profile holds conversational memory, the observations and summaries extracted automatically after each conversation closes, and factual memory, the traits and identifiers that describe the customer. New observations are added, merged with an existing one when they add detail, replace it when they contradict it, or are discarded when they lack support, so the profile stays current.

A Recall API returns the most relevant observations, summaries and recent messages by semantic and lexical search, and Agent Connect calls it on every messaging turn. Up to five stores separate data such as sandbox and production, each profile holds up to 500 observations and 500 summaries, and observations can be switched off per configuration. Memories carry no fixed retention period and persist until they are reconciled, edited or removed.

Sourcetwilio.com/docs/conversations/memory/conversation-memory-conceptsread 2026-10-11

Trust OS has kill switch and escalation controls that pause, redirect or kill any AI flow instantly, with one-click escalation to a human agent and role-based access control so the right people own the right agents. Policy is customer-authored through the Actionbook Editor, which lets operations teams edit refund thresholds, escalation logic and tone in real time with no code and no engineering ticket.

Hallucination and unsafe content detection runs automatically, flagging factual errors, off-policy responses and safety violations before customers see them, and a review-and-grade surface lets people score responses and flag outliers against the customer's own standards. Handoff to a person runs through the Zendesk, Salesforce and Freshworks integrations. Oversight comes through authored policy, automatic detection and interruption, not a sign off on each action before it is sent.

SourceDelight.ai Trust OS and AI agent platform pagesread 2026-09-05

The built in escalation tool hands a conversation to a Studio flow and on to a Flex agent with full context, on voice after the agent's last reply plays and on digital channels by posting straight to the flow. When and why to escalate is the customer's code to decide, and Agent Connect makes no decisions about what the agent does.

Conversation Memory applies a prompt exclusion meant to screen out GDPR special category data. Agent Identity adds scoped, revocable tokens and human approval over SMS or RCS before a high risk action, but it is in private release, so approval before each action is not yet open to every customer.

Sourcetwilio.com/docs/conversations/agent-connect/core-conceptsread 2026-10-11

SOC 2 Type 2, HIPAA, GDPR and CCPA are named on the security page, and ISO 27001 and HIPAA/HITECH on the platform page, with a live trust center at trust.sendbird.com.

Controls the customer sets include granular role-based access controls, full audit logs, flexible data retention settings aligned to its own policy, end-to-end encryption and TLS in transit, PII protection and data masking, and RBAC also scopes which people own which agents.

Internally, Sendbird uses FIDO2 multi-factor authentication with device trust, automated detection and response, automated compliance and a DevSecOps development approach, and it publishes a vulnerability disclosure address.

SourceDelight.ai security page, AI agent platform page and Trust OS pageread 2026-09-05

Twilio Console sign in works through a corporate identity provider over SAML 2.0, with SP and IdP initiated login and multiple SSO profiles, on Twilio Editions that include single sign on. Which accounts a user can open and which role they hold are set in the Console and Admin Center, and Flex maps identity provider roles to its own.

Twilio holds ISO/IEC 27001 certification and is audited externally on a regular cycle, with security and privacy reports available on request through security.twilio.com. Agent Connect and Conversation Memory are not HIPAA eligible or PCI compliant, so regulated workflows sit outside them.

Sourcetwilio.com/docs/iam/single-sign-onread 2026-10-11

Trust OS logs everything for tracing, and Openbook logs every conversation, every decision and every tool call for review, with no black boxes. Openbook also renders the agent's actionbook as a visual flow, so an operator can follow the logic behind a decision. Low CSAT Analysis sorts every conversation scored one or two into one of five failure types before anyone opens it, and Thinking Messages show a status for each tool call during the run, so the reasoning is visible as it happens as well as afterward. A dedicated Analytics product tracks conversations, common topics and ROI.

SourceDelight.ai Trust OS and Analytics pagesread 2026-09-05

Conversation Orchestrator stores every message and transcribed call as a threaded conversation linked to a profile, Conversation Memory shows the observations and summaries it extracted, and Conversation Intelligence analyzes each conversation for sentiment, summaries and custom criteria. Agent Connect logs a warning when it falls back from the Memory API. Together that shows what was said and what the system learned. The tool calls and reasoning live in the customer's agent, so a step by step trace of a run comes from the customer's own logging.

Sourcetwilio.com/docs/conversations/memoryread 2026-10-11

Customers keep full control over where their data resides, how it is stored and who can access it, to meet regulatory and internal data governance requirements, with regional hosting to support data residency needs. The platform lists region-specific storage among its privacy-first data governance properties, alongside PII protection and data masking. Retention is customer-set, with flexible data retention settings aligned to the customer's own policy, and data ownership and processing are described as transparent. No on-premise or self-hosted deployment is offered.

SourceDelight.ai security and AI agent platform pagesread 2026-09-05

The agent and its model run wherever the customer hosts them, since Agent Connect is an SDK inside the customer's own application. Twilio's isolated regions are US1 in the eastern United States, IE1 in Dublin and AU1 in Sydney, and a customer chooses one to keep a workload's data there.

Voice, SIP trunking and messaging run in Dublin, and voice in Sydney, while Conversation Memory, Orchestrator, Enterprise Knowledge and Conversation Relay are offered in the US region only, so the agent layer runs in the US. While regions roll out, Twilio does not guarantee that all data stays in the selected region.

Sourcetwilio.com/docs/global-infrastructure/regional-product-and-feature-availabilityread 2026-10-11

Capabilities each have their own page and are adopted as units, among them Agent Steward for autonomous ownership of long-running issues, Agent Memory Platform, For You Conversations, Omnipresent, Voice AI Agent, Trust OS, Analytics, Desk and AI Agent Integrations. Industry pages cover retail, travel, healthcare, financial services, logistics and more.

There is no template gallery, agent library or marketplace for customers to browse and put into service, and agents are built by onboarding them like a new team member, transferring knowledge and providing workflows through a supervisor dashboard, with behavior authored in an actionbook.

SourceDelight.ai AI agent builder, capabilities index and industry pagesread 2026-09-05

Agent Connect starts from a quickstart and integration guides for AWS Bedrock and Microsoft Foundry, Conversation Intelligence ships Twilio authored language operators alongside custom ones, and Twilio Skills and the Code Exchange give developers guided patterns and sample apps. Each of these is a starting point the customer builds on. Twilio does not offer finished agents for support, sales or scheduling that a customer switches on.

Sourcetwilio.com/docs/conversations/agent-connectread 2026-10-11

Delight.ai does not say which models run the agent or who chooses them, and there is no model selection, routing configuration or bring-your-own-key path. The product is sold as a managed outcome, with resolution rates, a governance layer that detects hallucinations, and zero-touch improvement that tunes the agent on its own.

SourceDelight.ai AI agent platform, Trust OS and security pagesread 2026-09-05

The customer picks the model. Agent Connect does not run inference. The customer supplies the LLM application and controls how it responds, with integration guides for AWS Bedrock and Microsoft Foundry and support for OpenAI or any other provider. Tools are formatted for whichever provider the customer uses, so switching models is a change in the customer's own code. Twilio's own generative features, such as observation extraction in Conversation Memory, run on models Twilio chooses.

Sourcetwilio.com/docs/conversations/agent-connectread 2026-10-11

Sendbird publishes AI agent documentation at sendbird.com/docs/ai-agent, linked from delight.ai's developer menu alongside a public community forum and a server status page, plus a GitHub organization at github.com/sendbird.

The underlying communications platform ships cross-platform SDKs for iOS, Android, JavaScript, React Native and Unity with chat, voice and video APIs, which embed agents natively in a customer's own app instead of as a widget.

AI Agent Integrations connect the agent outward to Zendesk, Salesforce and Freshworks, and Trust OS exposes the actionbook as an editable artifact that operations teams author directly. There is no MCP server or agent manifest.

SourceSendbird AI agent documentation, delight.ai developer menu and the Sendbird communications API platform; sendbird.com/docs/ai-agent/guide/v1/overviewread 2026-09-05

Every product has a public REST API and helper libraries, including a Conversation Memory API with profile, trait, observation, summary, import and Recall endpoints. Agent Connect itself is an SDK for Python and TypeScript with a FastAPI server for webhooks and WebSockets. An MCP server gives coding agents semantic search across more than 1,500 endpoints in over 30 products, and Twilio Skills package product guidance for AI coding tools.

Sourcetwilio.com/docs/airead 2026-10-11

Every change is tested before it touches a customer, with staging, real-conversation test suites, gradual rollout and A/B data, so a deploy comes with evidence. Staging environments mirror production with real integrations and policies. Test Upgrade builds test cases from real conversation IDs, organizes them with labels, shares sets across environments and returns consistent pass/fail results.

Gradual rollout ships to ten percent first, and A/B testing runs two versions head to head for hard data on resolution rate and CSAT before committing. Simulated environments test and evaluate an agent before deployment, and the Trust OS review-and-grade surface gives every response a quality score.

SourceDelight.ai Trust OS and AI agent platform pagesread 2026-09-05

Conversation Intelligence runs language operators over voice and messaging conversations, Twilio authored ones for sentiment and summaries and custom ones written as natural language prompts, which a team can point at its AI agent's conversations to flag compliance misses or missed upsells. That reviews live traffic after the fact. The Agent Connect quickstart builds and tests an application locally, and there is no suite of saved test cases with pass or fail results to run against a new version of an agent before it goes live.

Sourcetwilio.com/en-us/products/conversational-ai/pricingread 2026-10-11

The agent acts through integrations with Zendesk, Salesforce and Freshworks, through tool calls that Trust OS logs individually, and through the communications APIs and SDKs that embed it in the customer's own app. It talks to people over in-app chat, web, email, SMS, WhatsApp, social and voice, and Delight.ai has no browser control, hosted session, virtual desktop or interface automation.

SourceDelight.ai AI agent platform and Trust OS pagesread 2026-09-05

The agent layer connects an agent to communication channels, customer memory and knowledge through APIs and tools the customer writes. No hosted browser, remote desktop or interface control is part of it. An agent that has to work inside a web page or a desktop application brings that capability as a custom tool in the customer's own code, outside anything Twilio runs.

Sourcetwilio.com/docs/airead 2026-10-11