Embedded AI Backends

Ship AI features by exposing workflows as APIs, MCP servers, and portals instead of building new backend services.

The problem

Shipping one AI feature usually means standing up a service: orchestration, retries, provider SDKs, auth, and deployment. The feature is small, the infrastructure is not.

How it runs today

Product teams hand-roll framework code and glue services for every feature, then keep maintaining all of it as models, providers, and requirements change.

With Heym

Build the behavior as a workflow, then expose it directly as a token-protected JSON API, an MCP server, or an authenticated portal with structured responses.

The workflow, end to end

The Async Sub-workflow Dispatcher template is an importable starting point: it validates an incoming request and dispatches a reusable sub-workflow in the background, the pattern behind long-running workflow APIs. This website runs its own contact form on a Heym workflow in exactly this way.

Product or client requestWorkflow API or MCP callAgents, logic, and integrationsStructured response
Async Sub-workflow Dispatcher
View template

See it in Heym

A short walkthrough recorded in the product.

Exposing workflows as an MCP server so Claude, Cursor, and other clients can call them as tools.

Where control lives

Bearer token auth on execute endpoints, portal-level authentication for end users, and full run history for every external call.

AI features in production without owning another service: one workflow serves your product, your agents, and outside tools.

Built with these Heym capabilities

Webhook workflow APIs
Bearer token auth
MCP server exposure
Authenticated portals
Structured JSON outputs
Async sub-workflow dispatch
Execution history
Provider-agnostic models

Common applications

JSON APIs for product featuresContact and intake form backendsMCP servers for agent clientsAuthenticated chat portalsAsync job processingInternal tool backends

Deployment and integration

  • Self-host with Docker or Kubernetes, keeping data on your infrastructure.
  • Connect your own model providers: OpenAI, Ollama, vLLM, and more.
  • Integrate over HTTP, webhooks, Slack, email, and MCP tools.
  • Expose finished workflows as APIs, portals, or MCP servers.

Have a process that fits this pattern?

Show us the workflow, data sources, tools, and human decisions involved. We will help map it to Heym.