What we measure, how we measure it, and what gets tested as a result.
Active pilot · on-prem deployment ready · Mode 2 (integrator)
Replaces ad-hoc QA spreadsheets with a structured layer that plugs into your existing Redmine / JIRA / Azure DevOps, adds test design, execution capture, and evidence chains — and reports that survive audits.
Public · GitHub
Research code exploring kinetic + heat propagation across five language backends (Rust, Scala, C++, Fortran, Pascal). Used as a reference vehicle for cross-language CI patterns and numerical-validation workflows.
View on GitHub → LinkedIn post →
Public · GitHub
Standalone demos and snippets supporting the work above — CAST framework explorer, measurement-story examples, integration fragments. Use them as starting points for your own pilots.
View on GitHub →
Coming soon
An interactive walkthrough of the CAST 2.4 framework that motivates MI-M-T's data model. Publication pending — early v0.3 deliverable.
The work above sits at the intersection of measurement discipline and software craft. Each pilot follows the same four-step pattern: Impulse → Test Set → Execution → Evidence — adapted to the domain (test management, physics simulation, framework didactics).
MI-M-T is a meta-informed measurement layer for testing at every level: manual, scripted, recorded, and (in research mode) LLM-driven. It complements existing trackers (JIRA, Redmine, Azure DevOps) and adds structured test design, execution evidence, and traceability chains. Three deployment modes — replacement, integrator, and a research mode for deterministic LLM-driven TDD — share the same data model.