Interlock

A licensed integration platform, in two halves.

Interlock Studio is where an integration is discovered, built, and tested. Interlock Connect is the connector library it produces, running on whatever contemporary Java backbone you already operate. License either. License both. Run them with your engineers or ours.

The bifurcation

Two products, because they solve two problems and rarely arrive together.

Most integration tooling forces one purchase: adopt the vendor’s runtime and you get their tooling, or adopt their tooling and you inherit their runtime. Interlock separates the two deliberately.

Studio is a workbench - discovery, development, and test. It produces artifacts, not lock-in. Connect is 35 maintained connectors written against a thin adapter interface, so they deploy onto the Java backbone your platform team has already standardised on.

Interlock Studio

Discovery, development, and testing in one workbench - so the integration is specified before it is written.

Discovery

Studio ingests API documentation, schemas, and sample payloads and drafts the system map, dependency picture, and a risk score against a pattern library built from real engagements. Output is a written Architecture Brief, engineer-reviewed and signed.

Development

The discovery output becomes the input specification a connector is configured against: field mappings, transformation rules, and flagged edge cases - nullable fields, type mismatches, custom picklists. Specifications are versioned artifacts, reviewable in your own pull requests.

Testing

Contract tests generated from the specification, replayable fixtures captured from live traffic, and schema-drift detection that fails a build when an upstream system changes shape. Test suites are yours and run in your own pipeline.

AI sits in exactly two places inside Studio: carrying out system integration discovery, and defining the inputs connectors run against. Every output is reviewed and signed off by an engineer before it reaches a build. What Studio does not do is write your integration for you - that is precisely the claim you should distrust.

Interlock Connect

Connectors written against a thin adapter interface rather than a proprietary runtime. Each one handles the unglamorous part - authentication, pagination, rate limits, retry semantics, idempotency, error taxonomy - and deploys onto the backbone you already run.

01

Runtime-neutral by design

Certified against Apache Camel, Spring Integration and plain Spring Boot, Quarkus, and Jakarta EE application servers, on JDK 17 and above. A connector is a library on your classpath, not an appliance in your network.

02

Your deployment topology

Run them in-process inside an existing service, as standalone microservices, or as sidecars in your cluster. Interlock has no opinion about your orchestration and does not require an eProxim-hosted control plane.

03

Maintained against upstream drift

When a vendor deprecates an API version or changes a schema, the connector is updated and released. That maintenance is the substance of the subscription - the code you already have keeps working.

04

Source-available under licence

Licensees receive source. Your team can read it, debug it, and patch it in an emergency without waiting on us. Nothing about Interlock stops you operating what you have bought.

The connector library

35 systems, honestly labelled. Some are generally available and have carried production traffic for years; some sit in the library awaiting their first client environment; some we build to order and then maintain like the rest.

​Generally available​In the library​Built to order

CRM & Sales
Salesforce Generally available
HubSpot In the library
Microsoft Dynamics 365 In the library
Zoho CRM Built to order
ERP & Finance
SAP Generally available
NetSuite Generally available
Oracle ERP Cloud In the library
Sage Intacct Built to order
QuickBooks In the library
Coupa Built to order
Service & Workforce
ServiceNow In the library
Workday In the library
Jira & Jira Service Management Generally available
BambooHR Built to order
ADP Built to order
Commerce & Payments
Shopify In the library
Stripe Generally available
Adobe Commerce In the library
BigCommerce Built to order
Amazon Marketplace In the library
Braintree & PayPal In the library
Legacy & On-Premise
AS/400 (IBM i, DB2) Generally available
IBM Mainframe (MQ, CICS) Generally available
EDI (X12, EDIFACT) Generally available
SFTP & flat-file feeds Generally available
Oracle Forms Built to order
Data & Analytics
Snowflake In the library
Databricks Built to order
Amazon Redshift In the library
SQL Server & Oracle DB Generally available
Messaging & Middleware
Apache Kafka Generally available
RabbitMQ Generally available
Azure Service Bus In the library
MuleSoft In the library
Boomi In the library

Not listed is not a problem. Anything with an API, a message broker, a database, a file drop, or an EDI feed is in scope - the list reflects what has come up most, not the limit of what we connect. Commissioned connectors are quoted per system and maintained on the same terms as the rest of the library.

How it is licensed

Annual subscription, priced on environments and connectors rather than seats or message volume. No per-transaction metering.

Studio

Per-team annual subscription covering discovery, development, and test tooling, the pattern library, and generated contract-test suites. Artifacts you produce are yours outright and outlive the subscription.

Connect

Annual subscription per production environment, scoped to the connectors you select. Includes upstream-drift maintenance, security patches, and version upgrades for the term.

Commissioned connectors

A system not yet in the library is quoted once as a build, then folded into the maintained library at no additional subscription cost for that connector.

OEM and embedded

Partners embedding Interlock inside their own product or managed service license it under their brand, with redistribution rights and a named engineering contact. The end client need never hear our name.

Delivered with engineers

Any of the above can arrive with our team attached - fixed-scope build, embedded engineers, or architecture ownership only. The licence and the labour are priced separately, so you can take one without the other.

Fair questions

Do I have to buy both?

No, and most buyers start with one. Teams with a mature platform and a connector gap take Connect alone. Teams facing an unmapped estate take Studio first and license connectors once discovery says which ones they need.

Does this replace my integration platform?

No. Connect runs on the Java backbone you already operate - Camel, Spring, Quarkus, or an application server. Interlock is what you put on it, not a replacement for it.

What if we stop paying?

Deployed connectors and everything Studio produced keep running - you hold source under licence. What lapses is maintenance, upstream-drift updates, support, and new versions. No kill switch, no phone-home, no hosted dependency.

Who owns what?

Interlock is ours and licensed to you. Your mappings, specifications, tests, and integration logic are yours. Generic connector improvements funded by a commission go back into the maintained library, which is why that build is priced as a build and not as bespoke software.

Can we self-host entirely?

Yes, and most do. Studio and Connect run inside your own network and pipeline. There is no requirement to send code, schemas, or traffic to us.

Bring us a system pair and we will tell you whether a connector already exists for it.