Product
Intu.dev
Integration as Code for healthcare
intu.dev is an open-source, Git-native, AI-friendly healthcare interoperability framework built for teams that want version-controlled, reviewable integration pipelines. Channels are defined as declarative YAML with TypeScript transformers, stored in plain directories in Git, and executed by a fast Go engine. It handles everything from FHIR APIs to HL7 v2 interfaces without locking you into a proprietary platform.
Intu.dev is open source under the Mozilla Public License (MPL). Use, modify, and distribute it with confidence.
Integration as Code
YAML + TypeScript pipelines in Git. No GUI or database config.
Healthcare-Native
HL7 v2, FHIR R4, X12, CCDA, and DICOM built in.
25 Connectors
12 sources and 13 destinations: HTTP, MLLP, Kafka, SFTP, DICOM, and more.
Fast Go Engine
Sub-millisecond transforms, hot reload, Node.js worker pool.
Built-in Dashboard
Monitor channels, browse messages, trigger reprocessing.
Production-Grade
Redis clustering, retries and DLQ, OpenTelemetry, secret managers.
Why Git-Native?
Healthcare integrations are critical infrastructure. By treating pipelines as code in plain directories tracked in Git, you get collaboration through pull requests, full history and rollback, and CI/CD with the same workflows your engineering team already uses. No black-box config UIs, no database-resident configuration, no vendor lock-in; your integration definitions are yours, in your repo.
Healthcare-Native Connectors
Parsers and connectors for HL7 v2, FHIR R4, X12, CCDA, and DICOM are built in. Twelve source and thirteen destination connector types cover HTTP, TCP/MLLP, FHIR, Kafka, databases, files, SFTP, email, DICOM, SOAP, IHE, Direct, and more. The systems you need to reach are supported out of the box, whether they speak modern APIs or thirty-year-old protocols.
Fast and Production-Ready
A Go binary paired with a Node.js worker pool delivers sub-millisecond transforms and hot-reload during development. In production you get Redis-backed clustering, retries with dead-letter queues, OpenTelemetry and Prometheus observability, and secrets management via Vault, AWS, or GCP. A built-in web dashboard lets you monitor channels, browse messages, and trigger reprocessing, and a Mirth XML importer eases migration from legacy engines.
AI-Friendly by Design
Structured schemas and declarative config make it straightforward for AI assistants to generate and modify pipelines. Whether you use an LLM to draft a new channel or to refactor an existing one, intu.dev's format is predictable and easy to reason about, which reduces errors and speeds up development.