xF is a metadata-interpreted runtime for building and running enterprise business systems. Nothing is generated, nothing is compiled, and the design can never drift from the running application.
Not low-code. Not a code generator. Not a workflow engine. No-build.By separating the intent of software from its execution, xF gives business systems a foundation that changes as fast as the business does.
Model-driven tools, low-code builders, API frameworks and workflow engines each solve a slice of the problem. Teams then spend years keeping the slices in sync.
xF treats the whole path, from schema to API to screen to AI assistance, as one thing.
What most stacks assemble and keep in sync by hand.
Define a concept once. Storage, API, validation, screens and business rules all read it live, so a change is a single edit that takes effect on the next request.
One architectural decision, followed all the way through: keep the model alive as data instead of freezing it into code.
Entities, fields, controls, routing and behaviour are defined once, as data. The physical schema is maintained for you, and xF is administered with xF itself, demonstrating recursion in practice.
Four generic endpoints serve every entity and every workflow. The domain grows; the API surface does not.
Each request carries its own intent and state. No server sessions, and writes are version-checked before commit.
Queries live in the registry. Save a row and a new report or dashboard is live, with no build cycle.
Screens resolve from the same control types the model defines. Use the native renderer, or go headless with React, Vue or mobile.
Access rules are declared once, down to individual fields, and hold everywhere. "Only HR can see salary" is a metadata rule, not custom code.
Routine behaviour is automatic CRUD, custom rules run as sandboxed C# or VB.NET, more complex logic can be deployed as compiled algorithm assemblies, and anything else routes out to external services. All four share the same execution model.
MCP is an open, cross-vendor standard that lets AI tools discover a product, read its structure and use its capabilities. Paired with xF's deterministic runtime, it gives enterprises a governed way to build and evolve software with AI, without piles of generated code to maintain.
MCP gives the AI a controlled interface and xF gives it a deterministic runtime. The AI touches metadata, never arbitrary code.
The MCP server exposes the meta-model, documentation, optimisation patterns and samples, so the AI learns xF from xF itself.
Model the application, then the logic, then the frontend. Each step is a narrow target, which keeps output accurate by construction.
Changes are small, readable metadata rather than sprawling generated code, so long-term ownership stays predictable and cheap.
An open standard already used by tools such as Claude, ChatGPT, VS Code and Cursor. Use the assistant your team prefers.
Plain-text metadata, standard SQL and an open frontend. MCP is open too, so there is no proprietary trap.
Real systems under real load. Details shared under NDA, and several have been independently evaluated.
Malta. Bookings, payments, field operations.
More than five years of operation.
Built for a 20-bus operator. A concrete, inspectable example of how the approach scales.
Also delivered on xF: quantity surveying and cost control, multi-tenant time and billing SaaS, and a specifications database for a large residential programme.
Three ways in, built on the same runtime. We onboard people personally, one at a time, so there is no self-serve sign-up. Tell us what you want to build and we will set you up.
For developers, evaluators, students and agencies prototyping.
For teams launching live products without managing infrastructure.
For enterprises and regulated sectors that need full sovereignty over their runtime.
xF is built so that you can leave. That is also why it is easy to trust.
Your data sits in your database, in a schema you can read. The metadata is plain data too.
The interface layer is open, so screens and controls can be extended or replaced.
Run it on your own infrastructure, on premise or in your cloud.
The architecture behind xF, and how it separates the intent of software from its execution, is described in our technical paper.
Read the technical paperWe work with partners who lead enterprise transformation, systems integration and legacy modernisation initiatives. If your organisation delivers mission-critical platforms and is looking for a fundamentally different runtime approach, we'd welcome a conversation.
Email the teamWe partner with Microsoft Clarity and Microsoft Advertising to capture how you use and interact with our website through behavioral metrics, heatmaps, and session replay to improve and market our products and services. Website usage data is captured using first and third-party cookies and other tracking technologies to determine the popularity of products and services and online activity. We also use this information for site optimization, fraud and security purposes, and advertising. For more information about how Microsoft collects and uses your data, visit the Microsoft Privacy Statement.