AI & automation
Amazon Lex, Google Dialogflow & Nuance bots
Run Amazon Lex, Google Dialogflow, Nuance and other bot platforms inside Genesys Cloud CX, with interface contracts, routing and handover owned by one certified QVCCS team.
A Genesys Cloud third-party bot integration lets you keep the conversational AI you have invested in and run it inside your Genesys contact centre. QVCCS connects Amazon Lex V2, Google Dialogflow CX and ES, and Nuance Mix bots through Genesys-supported integrations, and uses the Genesys Bot Connector for platforms such as Kore.ai or Cognigy. Specialists from our own bench write the interface contract, build the Architect flows that call each bot, prove every handover and failure path, and support the result end to end.
- Solution Architect
- Senior Developer
- Software Developer
- IT Systems, Telecoms & SIP Engineer
- Systems Integration Tester
- User Acceptance Test Lead
- Keep your bot investmentExisting intents, training data and bot teams carry forward, while Genesys Cloud CX handles channels, routing and agents.
- Supported, documented integrationsAmazon Lex V2, Dialogflow CX and ES, and Nuance Mix connect through AppFoundry integrations called from Architect actions.
- Room for other platformsThe Genesys Bot Connector brings up to five third-party bots into message flows through a translation layer we build.
- One journey for customersAn interface contract governs every variable crossing between bot, flow and agent, so customers never feel the join.
- Nuance Mix end of support on Genesys Cloud: what ends and your optionsGenesys ends support for Nuance Mix bots, Nuance Recognizer and Nuance TTS on 31 May 2027. What is affected, how to find it, and the options that remain.Read the guide
- Voice AI cost calculator for Genesys Cloud: native and third-party comparedCompare the list-price cost of Genesys bot flows, virtual agents and agentic agents with Google, Amazon, Deepgram and ElevenLabs voice AI, per resolved call.Read the guide
- Native or third-party voice AI on Genesys Cloud: how to chooseGenesys bot flows, virtual agents and agentic agents, or Dialogflow, Lex, Deepgram, ElevenLabs and others? How each connects, costs and is governed.Read the guide
- Amazon Lex V2Call Lex V2 Bot action
- Dialogflow CXCall and message flows
- Dialogflow ESExisting Google agents
- Nuance Mix BotSupport ends May 2027
- Bot ConnectorKore.ai, Cognigy, others
- Agent handoverContext to queues
Architect orchestrates each bot, carrying data in and routing results to queues and agents.
01
Why a Genesys Cloud third-party bot integration
A Genesys Cloud third-party bot integration makes sense when your organisation has already built conversational AI on another platform, or has a strategic reason to standardise on one. Perhaps a digital team runs bots on Amazon Lex for your website and app, or a group-wide AI programme has chosen Google Dialogflow. Rebuilding those bots natively may be unnecessary. Integrating them with Genesys Cloud CX lets your bot team keep its tools, training data and release process, while the contact centre gains a single place for channels, routing, agents and reporting.
The commercial and operational logic matters as much as the technology. Running a bot outside Genesys Cloud CX means two platforms to license, monitor and support, and two teams who must agree how changes are made. In Discover, QVCCS audits your bot estate – platforms, channels, owners, release cycles and performance – and in Design we record a formal options analysis weighing the existing investment against native Genesys bot flows and AI Studio. Each recommendation goes into the design decision log, so the reasoning survives long after the project team has moved on.
02
Supported integrations and the Bot Connector
Genesys supports integrations with Amazon Lex V2, Google Dialogflow CX, Google Dialogflow ES and Nuance Mix, each installed from AppFoundry, configured with provider credentials and called from a dedicated Architect action. The details differ. Lex V2 bots are called from call flows and are documented as PCI DSS compliant for secure flows. Dialogflow CX bots can be called from call and message flows and are also documented as PCI DSS compliant for secure call flows, but are not supported on BYOC Premises or premises-based Edge. Dialogflow ES bots work in call, chat and message flows, but Genesys advises against using them in secure call flows because they are not PCI compliant.
Two status changes need planning. Nuance has announced end of life for Nuance Mix Dialog, and Genesys will discontinue support for the Nuance Mix Bot integration on 31 May 2027, so we help Nuance customers plan a move to another platform or to native bot flows. For other platforms, the Genesys Bot Connector lets you add up to five third-party bots to Architect message flows through a translation layer you host. Genesys has also documented the Genesys Digital Connector, currently shown as coming soon, for asynchronous sessions lasting up to 72 hours, and now labels the Bot Connector as legacy and recommends the Digital Connector for new integrations. Kore.ai and Cognigy both document Genesys Cloud integrations.
Customers should never feel the join between your bot platform and your contact centre – the interface contract is where we make sure they don't.
03
Integration design decisions that protect the customer
Two platforms mean two places where a conversation can break, so the interface needs precise definition. QVCCS writes an Interface Control Document for every bot integration covering session start and end, the variables passed in such as caller identity, language, channel and authentication state, the values returned such as intent, slots and outcome codes, and the behaviour expected on timeout, error or an unexpected response. That contract becomes the shared reference for building and testing on both sides, and the first document an engineer reaches for when diagnosing an issue in live service.
Voice deserves special attention. Speech recognition, prompts, barge-in and silence handling can sit with the bot platform or with Genesys Cloud CX depending on the integration, and the choice affects latency and experience; for Dialogflow CX, for example, an Architect text-to-speech engine can override the bot's own voice. Our IT Systems, Telecoms & SIP Engineer confirms your telephony model is supported, response times are measured and prompts and timeouts are designed accordingly. We agree data residency and privacy positions with your security team, because customer utterances will also be processed by the bot provider.
Handover and reporting must join up. When a third-party bot decides a customer needs a person, the Architect flow should receive a clear reason and the captured details, write them into participant data, and route to the right queue with a summary for the agent. Reporting should show bot outcomes alongside queue and agent outcomes. Our developers define outcome codes, participant data and wrap-up conventions in the Low-Level Design, so your analysts can follow a customer from first utterance to resolution, whichever platform handled each step, and every convention is traced to a requirement in the RTM.
04
How QVCCS delivers third-party bot integration
Third-party bot integration sits between two teams, so QVCCS musters a team from its own bench that can work credibly with your bot team or vendor. A Solution Architect owns the High-Level Design and the interface contracts, and is the person your bot provider agrees each clause with. A Senior Developer and Software Developer configure the integrations, build the Architect flows and any Bot Connector translation layer you host. An IT Systems, Telecoms & SIP Engineer covers voice paths, where latency and recognition are decided. A Systems Integration Tester and User Acceptance Test Lead prove the result, and the Senior Platform Practice Lead peer reviews the design. We take collective responsibility for the Genesys side and the interface between the platforms.
Integration quality is proven, not assumed. Our test strategy maps every ICD clause to a SIT test case in the traceability matrix. The SIT pack covers happy paths, every handover reason and language variation, while negative testing deliberately triggers provider timeouts, credential failures, malformed responses and missing variables to prove that the flow recovers gracefully. Voice journeys are tested on live telephony for latency and recognition quality. Agents run UAT scenario scripts to confirm the handover context is useful, and a cut-over runbook with a rollback plan governs the switch, with both bot and contact centre teams briefed.
05
Supporting a multi-platform conversational estate
Once live, the integration needs joint ownership. QVCCS writes an operational handover pack that sets out who fixes what across your bot team or vendor and the contact centre, so incidents are triaged quickly rather than bounced between teams. The signals worth watching are rising error rates, slow bot responses and changes in the mix of handover reasons, which can reveal a change on the bot side before anyone reports it. Release readiness reviews cover both platforms, including vendor deprecations such as Nuance Mix, and support follows your Genesys Cloud CX support model. If your strategy changes, we plan a phased move between third-party and native bots without disrupting service.
What you get from QVCCS
- Bot estate audit and options analysis for native versus third-party
- Interface Control Document for every bot integration
- Lex V2, Dialogflow or Nuance Mix integrations configured and secured
- Bot Connector translation layer designed, built and hosted where needed
- SIT pack and negative tests traced to every interface clause
- Nuance Mix exit plan ahead of the May 2027 support end
- Operational handover pack and support fitted to your provider model
Genesys documentation references
- About the Amazon Lex V2 integrationhelp.genesys.cloud
- Google Cloud Dialogflow CX integration overviewhelp.genesys.cloud
- About the Google Cloud Dialogflow ES integrationhelp.genesys.cloud
- Deprecation: Nuance Recognizer as a Service, Nuance TTS, and Nuance Mix Bothelp.genesys.cloud
- About Genesys Bot Connectorhelp.genesys.cloud
- Introducing Genesys digital bot connectorhelp.genesys.cloud
- Kore.ai – Adding Genesys Cloud CX Messaging as a channeldocs.kore.ai
- Cognigy – Genesys endpointdocs.cognigy.com
Methods & templates
Amazon Lex, Google Dialogflow & Nuance bots: the methods and templates we use
Every QVCCS engagement follows our seven-stage delivery lifecycle, each stage closed by a quality gate. For Amazon Lex, Google Dialogflow & Nuance bots, the work concentrates in Discover, Design, Prove and Transition, and these are the techniques and standard templates we lean on there, each one traceable from requirement to design, build, test and support.
- 01DiscoverDiscovery sign-off
- 02DefineRequirements baseline
- 03DesignDesign authority review
- 04BuildBuild complete
- 05ProveGo / no-go readiness
- 06TransitionOperational acceptance
- 07Run & evolveService reviews
Existing-org configuration audit
Inventories current bot integrations, flows, credentials and channels, including any Nuance Mix bots affected by the 2027 support end.
Options analysis and design decision log
Records why each journey uses a third-party bot, the Bot Connector or native bot flows, with trade-offs agreed.
Interface Control Document (ICD)
Defines session handling, variables in and out, outcome codes, timeouts and error behaviour between Genesys and each bot.
Negative and failure-path testing
Forces provider timeouts, credential failures and malformed responses to prove that Architect flows recover and route customers safely.
Cut-over plan and runbook
Sequences the switch to the integrated bot with rollback steps, monitoring checks and named owners on both teams.
How we deliver
Amazon Lex, Google Dialogflow & Nuance bots: who does what, and when
- 01DiscoverAudit your bot estate, owners, channels and telephony, then run an options analysis for each journey.
- 02ContractInterface Control Documents and designs agreed with your bot team or vendor before either side builds.
- 03IntegrateIntegrations, Architect flows and connector services built to the LLD with least-privilege credentials and peer review.
- 04ProveSIT, negative, telephony and UAT testing traced to the RTM, then a rehearsed cut-over.
- 05Support jointlyAgreed triage with your bot and Genesys Cloud CX providers, aligned support and release readiness reviews on both platforms.
- Solution Architect
- Senior Developer
- Software Developer
- IT Systems, Telecoms & SIP Engineer
- Systems Integration Tester
- User Acceptance Test Lead
Every engagement follows our seven-stage method, with design authority, engineering standards and four-eyes peer review behind it. How we deliver →
Questions
Amazon Lex, Google Dialogflow & Nuance bots: common questions
Which third-party bots integrate with Genesys Cloud?
Genesys supports integrations with Amazon Lex V2, Google Dialogflow CX, Google Dialogflow ES and Nuance Mix, each called from an Architect action. Other platforms can join message flows through the Genesys Bot Connector, and vendors such as Kore.ai and Cognigy document their own Genesys Cloud integrations.
How many bots can the Genesys Bot Connector support?
Genesys documents up to five third-party bot integrations with the Genesys Bot Connector, used in Architect message flows through a translation layer that you host. Genesys has also documented the Genesys Digital Connector for longer asynchronous sessions, currently shown as coming soon, and recommends it for new integrations once available.
Is the Nuance Mix Bot integration being retired?
Yes. Following Nuance's end-of-life announcement, Genesys will discontinue support for the Nuance Mix Bot integration, Nuance Recognizer as a Service and Nuance TTS on 31 May 2027; use beyond the end of 2024 depends on a service extension agreed with Nuance. QVCCS assesses your Nuance bots and plans a tested migration to another supported platform or to native Genesys bot flows well before that date.
Can a third-party bot pass customer details to a Genesys agent?
Yes, when the integration is designed for it. The bot returns intent, captured values and an outcome to the Architect flow, which writes them into participant data and routes to a queue. We specify that handover in the interface contract and prove it in UAT, so agents see useful context immediately.
Last reviewed