Skip to main content
EDD — home

Track record

Industrial software, delivered with EDD.

Real projects, built with the method and already running in production.

Read the method

Evidence

Accepted code, month by month.

Lines written by Claude Code and kept by the developers — “accepted” in Claude Code Analytics’ terms — on projects run with EDD, July–October 2026. A measure of output; the quality evidence is the check at the end of every phase.

Accepted lines of code

Claude Code Analytics · July–October 2026

  1. 27,848 July
  2. 128,766 August
  3. 143,998 September
  4. 177,810 October

478,422 accepted lines in total, July–October 2026

3 developers

each on a live project built with EDD:

  • Cloud migration
  • Android industrial application
  • Data-fabric environment and big-data analytics

3 projects in production now

Case studies

Two industrial systems, end to end.

Predictive asset health for industrial process equipment

A simulator of industrial process equipment built on first-principles models of thermal and hydraulic behaviour, feeding a full asset-health pipeline and an AI layer that turns condition data into clear, prioritised recommendations.

Predictive asset health for industrial process equipment: architecture Physics models (Wolfram Mathematica, thermal and hydraulic) — anomalies → Data platform (Microsoft Fabric, ingest and analytics) — dashboards → AI layer (plain-English reports, data products) — insight → Operators (equipment condition, prioritised actions). anomaliesdashboardsinsight Physics models Wolfram Mathematicathermal and hydraulic Data platform Microsoft Fabricingest and analytics AI layer plain-English reportsdata products Operators equipment conditionprioritised actions
  • Mathematical models in Wolfram Mathematica detect performance degradation early.
  • Microsoft Fabric carries ingestion, transformation, analytics and dashboards end to end.
  • An AI layer writes natural-language reports and data products for operators — a picture of equipment condition with prioritised actions, not raw telemetry.

Android field application

A full-lifecycle Android application in Kotlin and Jetpack Compose for engineers inspecting industrial assets on rugged field tablets.

Android field application: architecture Field tablet (Android, rugged) — capture → Field app (Kotlin, Jetpack Compose) — store → Local database (Room, offline first, the source of truth) — when online → IoT platform (industrial, reliable sync). capturestorewhen online Field tablet Androidrugged Field app KotlinJetpack Compose Local database Room, offline firstthe source of truth IoT platform industrialreliable sync
  • Offline first: identification, template-driven capture, location creation, a structured inspection and certificate generation all run against a local Room database, the single source of truth.
  • The field journey never depends on a network.
  • Online, records synchronise reliably with an industrial IoT platform.

See how the method works.

One governed flow for every change: people decide, agents do, evidence proves it.

Read the method See the governed flow