EN | IT | ES | DE

Engine Evolution

MySirt evolves while preserving its model-centric execution core. Its foundational principles remain stable: formal definition, deterministic interpretation, system derivation, and structural regeneration.

Evolution takes place by extending the expressive capabilities of the DSL, the Model Execution Engine, the generation mechanisms, and the shared runtime components, while preserving architectural coherence and deterministic execution.

Foundational Principle

  • Describe a system formally
  • Interpret the model deterministically
  • Derive executable structures
  • Maintain structural coherence over time

This principle remains unchanged since the engine’s original conception.

Architectural State

  • Declarative DSL as the formal system specification
  • Model interpreter with structural validation
  • Deterministic system derivation layer
  • Rule execution embedded in structural generation
  • Coherent regeneration capability

Evolution Through Real Projects

The MySirt Execution Engine evolves through requirements that emerge from real generated systems.

A project may initially introduce a capability needed for a specific operational context. When that capability proves reusable across domains, it can be generalized and incorporated into the platform through coordinated extensions of the DSL, the Model Execution Engine, the generation templates, or the shared runtime components.

From that point onward, the capability is no longer a project-specific implementation. It becomes part of the standard modeling vocabulary and execution capabilities available to future systems.

From Specific Requirement to Platform Capability

  • A concrete requirement emerges in a generated system
  • The requirement is implemented and validated in its operational context
  • Its general applicability is assessed
  • The capability is formalized within the MySirt architecture
  • The DSL and Execution Engine are extended where required
  • The capability becomes reusable across future generated systems

MySirt therefore evolves by increasing what can be expressed and executed through the model, rather than by accumulating disconnected custom implementations.

Examples of Engine Evolution

  • Printable document templates
  • Security improvements derived from penetration testing
  • Multilingual system generation and execution
  • Notification and communication mechanisms
  • Runtime infrastructure enhancements
  • Reusable workflow and execution patterns

These examples represent capabilities that can move from an individual project need to a reusable part of the MySirt ecosystem.

AI-Assisted Model Authoring

Artificial Intelligence complements MySirt by assisting users in understanding and describing domain knowledge before it is formalized in the declarative model. It can help identify candidate entities and relationships, suggest behaviours, and draft business rules for subsequent validation and formalization.

Authoring Role

  • Domain understanding and structural analysis
  • Drafting of entities, relationships, behaviours, and rules
  • Assistance in refining model definitions
  • Support for producing clearer executable specifications

Architectural Boundary

  • AI assists modelling; it does not execute the system
  • The declarative model remains the Single Source of Truth
  • AI may propose model content; only the validated declarative model is interpreted and executed
  • The Model Execution Engine remains responsible for deterministic interpretation
  • Runtime structures are derived from the model, not from AI inference

AI may improve modelling productivity, but it does not replace the execution architecture. Determinism, consistency, reproducibility, and architectural coherence remain responsibilities of the MySirt Execution Engine.

Conceptual Integration Flow

Domain Knowledge → AI-Assisted Modeling → Declarative DSL → Model Execution Engine → Executable Information System

MySirt executes formal models, not probabilistic outputs.

Architectural Position

  • AI is an assistive layer for domain analysis and model authoring
  • The validated declarative model remains the Single Source of Truth
  • The Model Execution Engine remains stable, independent, and deterministic
  • No execution responsibility is transferred to AI
  • New reusable capabilities become part of the platform through controlled architectural evolution