Back to vendors
A

Assembled

Visit site
Entry priceNot public; custom pricing across Core, Pro, and Enterprise tiers, quoted through salesFull pricing detail

AI customer support orchestration platform that unifies autonomous AI agents across chat, email and voice, an agent copilot, and AI powered workforce management, connecting automation directly to staffing across a blended human, AI, and outsourced workforce.

Assembled, based in San Francisco, is an AI customer support orchestration platform that unifies AI agents, an agent copilot, and workforce management so support teams can run humans, AI, and outsourced partners as one operation. Founded in 2018 by brothers Ryan Wang, its chief executive, and John Wang, both formerly at Stripe, it is backed by investors including NEA, Emergence Capital, and Stripe. Its thesis is that the hard part of AI in support is no longer deployment but orchestration, knowing which work AI should handle, where humans add value, and how to staff a blended workforce as automation shifts volume day to day.

The platform has three connected parts. Autonomous AI agents resolve customer inquiries end to end across chat, email, and voice, answering questions, executing account level actions, following workflows, and staying within brand voice and compliance guardrails.

An AI copilot works alongside human agents inside existing ticketing and CRM systems, drawing on the knowledge base, documentation, and successful past tickets to surface knowledge, draft responses, recommend next best actions, and enable one click task execution. Underneath sits AI powered workforce management with forecasting, one click schedule generation, and real time visibility across human agents, AI agents, and outsourcing partners.

Support orchestration ties automation directly to staffing, scoring which case types to automate, flagging knowledge gaps, and rebalancing in real time with dynamic handoff logic based on capacity and demand.

Assembled integrates with existing support tools through one click connectors and publishes public API documentation that developers can read without an account, including an API for embedding its agents into a company's own product. It is SOC 2 Type II certified against the 2017 Trust Services Criteria, HIPAA compliant with business associate agreements available, and GDPR compliant with a current EU-U.S. Data Privacy Framework certification, alongside annual third party penetration testing, AES-256 encryption at rest, and single sign on.

Its trust center, holding audit and penetration test reports, is available on request rather than published openly. It runs as a cloud service, and no customer choice of underlying model is offered. For a support or contact center team that wants automation and workforce management under one roof, with granular control across a blended workforce, Assembled is a strong fit; a team seeking only a lightweight standalone chatbot, or a model flexible platform, will find it broader than that.

Vendor details

Canonical URL

https://www.assembled.com

Category

Customer support agent

Subcategory

AI customer support orchestration and workforce management

Funding status

Independent, based in San Francisco, founded in 2018 by brothers Ryan Wang, chief executive, and John Wang, both formerly at Stripe. Assembled raised a Series B round with investors including Emergence Capital, Stripe, Felicis, Basis Set Ventures, and New Enterprise Associates. It serves support teams at companies such as ClassPass, ServiceTitan, Patreon, and Pair Eyewear, and positions itself as the platform that unifies workforce management with AI support automation.

Company status

independent

Use cases & customers

Primary use cases

AI customer support automationagent assist and copilotworkforce management and schedulingsupport orchestration

Target customers

customer support teamscontact centersBPO and outsourced support operations

Deployment options

SaaScloud

Integrations

Assembled connects to a company's knowledge, customer data, workforce data, and cases through one click integrations, works inside existing ticketing and CRM systems, and offers advanced APIs that let teams embed its AI agents directly into their own product. It is listed on the Salesforce AppExchange and its agents execute account level actions across connected internal tools and policies.

In practice

Your support volume swings daily and you cannot tell whether to add agents or lean on automation. Assembled forecasts demand, builds schedules in one click, and rebalances between human and AI agents in real time as queues run hot.

You deployed a support chatbot but have no idea which tickets it should handle or where it hurts quality. Assembled scores automation opportunities from your case history and routes work with dynamic handoff logic so quality holds.

New agents take weeks to ramp and quality is inconsistent. Assembled's copilot surfaces the right knowledge, drafts on brand responses, and enables one click actions inside your existing helpdesk to lift first contact resolution.

Agentic Index coverage score

8.5 / 14 capabilities · 61%

Integrations & Tool Calling Full

Named integrations with authenticated write action: Assembled publishes native integrations with Zendesk, Salesforce, Shopify and others that pull data and trigger actions automatically, and interactions are enriched with order history and account details drawn from those systems.

Agents execute account-level actions rather than only answering, and Copilot offers one-click task execution inside the customer's existing ticketing and CRM. A public API is documented at docs.assembled.com for embedding agents into the customer's own product, and one-click connectors are published on the integrations page.

Sourceassembled.com/features/agentic-workflowsread 2026-09-05

Workflow Orchestration Full

Multi-step execution by specialized agents rather than a single do-everything model: agentic workflows are published as multiple specialized AI agents handling multi-step, cross-channel journeys end to end, combining deterministic logic the customer can rely on with AI reasoning for the unpredictable parts, with the vendor stating each agent holds one task so accuracy holds across the chain. Above the individual case, support orchestration ties automation to staffing, scoring which case types to automate and rebalancing in real time with dynamic handoff logic across human agents, AI agents and outsourcing partners.

Sourceassembled.com/features/agentic-workflowsread 2026-09-05

Knowledge Grounding & RAG Partial

A shared knowledge layer is documented and its internals are not. Copilot serves agents policies, procedures and knowledge without their having to search documentation, and Assembled states that Copilot shares the same knowledge, workflows and data as the AI agents and the workforce management side, so one corpus serves the whole operation rather than each surface holding its own. Interactions are enriched with customer context pulled from connected systems.

No index, embeddings layer, retrieval architecture, refresh behavior or source-level citation is documented, so a maintained retrieval structure is asserted rather than shown, and no knowledge product page exists in the navigation to describe one. Knowledge-gap analysis is an automation-opportunity report rather than a grounding mechanism.

Sourceassembled.com/features/ai-copilotread 2026-09-05

Human Oversight & Guardrails Full

The gate is customer-defined and sits inside the flow. Assembled states that the customer defines the exact guardrails and policies, and that those guardrails include steps for human review or escalation placed within the workflow itself, so a sensitive path stops for a person rather than routing after the fact.

Agentic workflows are built to combine LLM reasoning with hard rules that enforce policy every time, giving deterministic control where it is needed, and brand style and company guidelines are configurable so responses hold to a set standard. Copilot carries the human side, guiding agents through cases that cannot be fully automated.

Sourceassembled.com/blog/agentic-workflowsread 2026-09-05

Security, Identity & Governance Full

Compliance and access controls are both documented, across three Assembled pages.

On compliance, the pages list SOC 2 Type II attested by The Cadence Group against the 2017 Trust Services Criteria, HIPAA compliance with Business Associate Agreements stated to be in place with every entity that could transfer protected health information and a BAA obtainable on request, GDPR adherence with a successful re-certification under the EU-U.S. Data Privacy Framework, an SSL Labs A+ rating, annual third-party penetration testing using dynamic fault injection and source-code review, secure development checks across the lifecycle, annual employee security training and a named incident-response channel.

A Vanta-hosted trust center at trust.assembled.com holds SOC 2 reports, penetration test reports and incident response plans; it is gated rather than open, with access granted manually on request. On access, the agentic workflows page names role-based access control and multi-factor authentication as enterprise security properties of the platform, alongside single sign-on, AES-256 at rest and SSL/TLS in transit, with employee access to customer data logged and audited. RBAC, MFA and SSO are controls the customer uses. ISO 27001 is not claimed.

Sourcesupport.assembled.com/hc/en-us/articles/40487316559629-how-does-assembled-keep-data-secureread 2026-09-05

Observability & Auditability Full

Reconstruction rather than reporting: the agentic workflows page publishes complete decision logs for every interaction, covering which data was used, what steps were taken, and why each path was selected, framed explicitly as full transparency into AI reasoning rather than a black box.

The vendor's own agentic-workflows post adds that every action and API call is logged and reviewable and that AI steps are explainable and traceable. Around that sit adoption, usage and impact metrics spanning agent performance, response quality and handle time, plus real-time visibility across human agents, AI agents and outsourcing partners.

Sourceassembled.com/features/agentic-workflowsread 2026-09-05

Memory & State Persistence Not documented

No memory layer with a stated scope or lifetime is documented. What exists is case continuity and retrieval: Copilot gives agents full case history and keeps context consistent when a case shifts teams, channels or time zones, and interactions are enriched with order history, account details and preferences pulled from the customer's connected systems at the moment of the interaction.

Both are reads against records the customer's own helpdesk and commerce systems hold, which is retrieval rather than state the agent retains, and a store is not a memory layer whoever writes it. No persistence a buyer could name, scope or delete is documented, and no memory product exists in the navigation.

Sourceassembled.com/features/ai-copilotread 2026-09-05

Deployment & Data Residency Not documented

Hosting region and residency options are not stated on Assembled's public pages. The security page covers encryption, access, penetration testing, incident response and secure development but names no region or cloud provider, and the help-center security and compliance article answers SOC 2, HIPAA and GDPR but not hosting location.

The trust center at trust.assembled.com is gated, with access granted manually on request within a few days, and holds SOC 2 reports, penetration test reports and incident response plans rather than a stated residency offering. The EU-U.S. Data Privacy Framework re-certification covers data transfer, not a location a buyer selects. Residency pages from similarly named companies, AssemblyAI, Assembly and Assembly Studio, do not describe Assembled. 2026-09-05

Prebuilt Agents, Templates & Packs Full

A named set of agents adopted by role and channel, each published with its own product page: AI Voice, AI Chat, AI Email and AI Copilot on the agent side, with AI-powered scheduling, ML-based forecasting, real-time analytics and vendor management as workforce modules alongside. A buyer selects the ones their operation needs rather than assembling each from parts. Custom agents are created rather than cloned, and no browsable template gallery or marketplace exists, so the adoption route is the named product set rather than a catalog.

Sourceassembled.com/features/ai-copilotread 2026-09-05

Triggers & Channel Coverage Partial

Three channels are documented and the agent receives rather than initiates. Assembled publishes AI Voice, AI Chat and AI Copilot as named agent products alongside AI Email, so work reaches the agent by phone, chat and email, and agentic workflows are described as handling cross-channel journeys.

No outbound or proactive trigger is documented, so the agent answers what arrives rather than starting a conversation. SMS does not appear in the product navigation, where the named agents are Voice, Chat and Email only. Real-time handoff logic and SLA alerts responding to queue demand are workforce-management signals inside the customer's own operation rather than triggers reaching the agent.

Sourceassembled.com/features/ai-voice-agentread 2026-09-05

Model Flexibility & Routing Not documented

No model selection, routing or bring-your-own-model control is exposed to the customer. Assembled's own site currently promotes a webinar on building a Claude skill, and a model provider's customer story describes Assembled Assist running on Claude and switching providers within twenty minutes during another provider's outage, which points to a third-party frontier model behind the product that the vendor can change at its discretion. What is absent is customer choice, not vendor-side flexibility.

Sourceassembled.com/securityread 2026-09-05

APIs, SDKs & MCP Extensibility Full

The platform is callable from outside through a documented public interface: Assembled publishes API documentation at docs.assembled.com, linked from the site navigation under Product Help and again from the footer, so a developer can read the surface without a sales conversation or an account. An integrations surface is published alongside it.

Sourcedocs.assembled.com/#introductionread 2026-09-05

Testing, Debugging & Optimization Partial

Measurement after deployment is documented; a gate before it is not. Assembled tracks adoption, usage and impact across agent performance, response quality and handle time, reports automation impact, and scores which case types to automate, so a change can be judged against readable outcomes once live. No simulation, scenario set, regression suite or scored evaluation of a proposed agent change before customers see it is documented. Scenario testing of schedules and forecast accuracy reporting exercise staffing plans and forecasts, not the agent.

Sourceassembled.com/features/ai-copilotread 2026-09-05

Browser & Computer Use Not documented

No browser control, hosted session, virtual desktop or computer use appears anywhere on the site. Instead, agents act programmatically, through native integrations with Zendesk, Salesforce, Shopify and others that pull data and trigger actions, through account-level actions inside connected systems, and through a documented public API.

Sourceassembled.com/features/agentic-workflowsread 2026-09-05

The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded

Recent platform changes

2026-10-01·Workflow orchestrationVerified

Assembled introduced Dynamic Routing for its AI chat agent, ordered rules that send a handoff to a specific queue or straight to email based on customer segment, language or topic. When a rule picks a queue, that queue's business hours in Assembled's workforce management decide whether the customer is offered live chat or email, and each routed conversation records which rule and schedule decision fired.

Bears on: Human approval / guardrails

View source
View all 1 change for Assembled →Tracked since Oct 2026 · Verified from public vendor sources

Pricing

Not public; custom pricing across Core, Pro, and Enterprise tiers, quoted through sales

tiered subscription (Core, Pro, Enterprise), quoted through sales

What is public

Core, Pro, and Enterprise tiers are named, with advanced forecasting reserved for Pro and Enterprise, but prices are quoted through sales and not listed publicly.

Billing mechanics

Tiered subscription across Core, Pro, and Enterprise, quoted through sales, with advanced forecasting features gated to Pro and Enterprise and pricing scaled to seats, channels, and volume, though rates are not disclosed.

Cost watchouts

Advanced forecasting sits in Pro and Enterprise, so the entry Core tier may not cover full workforce management, and cost scales with agents, channels, and volume across the blended workforce.

Variable cost rationale

Priced by tier and scaled to seats, channels, and AI agent resolution volume across a blended workforce, so cost grows with support scale and how much automation is deployed.

Additional watchouts

Confirm which tier includes the forecasting and orchestration features you need, and how AI agent resolution volume and seat counts scale the quote.

Sales call required

Yes, required for paid access

Free / trial

Demo on request; no public free tier

Lowest paid plan

Core tier (pricing not public), with Pro and Enterprise above it

Key ambiguities

Tier names are public but prices are not, and how AI agent usage, seats, and workforce management modules combine into a quote is not disclosed.

Agentic Index verified 2026-07-07

Alternatives to Assembled

The closest documented capability profiles to Assembled among customer support agents tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.

  • Fini8.5 / 14Fuller documented coverage on Knowledge Grounding & RAG and Triggers & Channel Coverage
  • Glia10.5 / 14Adds documented Memory & State Persistence and Deployment & Data Residency
  • LivePerson8.5 / 14Fuller documented coverage on Triggers & Channel Coverage and Testing, Debugging & Optimization
  • Parahelp8.5 / 14Fuller documented coverage on Knowledge Grounding & RAG and Testing, Debugging & Optimization
  • Tidio9.5 / 14Adds documented Deployment & Data Residency
  • Zingtree9.5 / 14Adds documented Memory & State Persistence and Deployment & Data ResidencyAssembled vs Zingtree →

Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded

Contact us

Found a vendor we missed? Have feedback on the index? We'd love to hear from you.