GitHub
New to FDLC? Start with the Guide.Start here

Factory Development Lifecycle (FDLC)Open framework · v0.1

Build the factory that builds your software.

The operating model and technical foundation for AI software factories that turn intent into verified software.

Start reading the GuideUnderstand FDLC in 5 minutes
See a factory in action →

For engineering leaders, platform teams, software-factory architects, and builders evaluating governed AI delivery.

SDLC

Governs the lifecycle of software.

FDLC

Governs the factory that builds it.

Code generation is becoming abundant. Trusted autonomous software delivery is not.

FDLC / Category promise

Verified delivery, not merely generated code.Autonomy without evidence is just faster guessing.

The FDLC governing lifecycle

Follow the work. Inspect the controls.

Select a stage to see what enters, what leaves, and what permits the work to move forward.

Stage 01 / 09

Intent

State the desired outcome, constraints, and accountable owner.

Input
A problem worth solving
Output
Intent contract
Control at this stage

Agree on scope, acceptance criteria, risk, and ownership before delegating work.

Example · dependency upgrade

Upgrade a vulnerable dependency. Preserve the public API and existing service behavior.

Learning closes the loop. Improvements return through verification and approval. This explorer illustrates the lifecycle; it does not execute work or grant authority.

Go deeper

Follow the complete guide.

Move from the factory overview to the chapters on design, agents, evidence, operations, and improvement.

Browse all chapters →

One illustrative example

Modernize a dependency without surrendering control.

A software factory does more than generate a patch. It coordinates bounded work, returns evidence, preserves human authority, and produces a candidate that can earn release.

  1. 01
    Intent

    Upgrade a dependency across approved repositories without changing public behavior.

  2. 02
    Factory

    Plan → Modify → Test → Verify → Approve

  3. 03
    Evidence

    Compatibility · Tests · Security · Cost · Human authority

  4. 04
    Outcome

    A production-ready modernization candidate with an attributable evidence trail.

Follow the complete exampleBuild your Factory Definition

Interactive factory explorer

Follow one unit of workOpen on Framework →

Choose a stage. Then inspect the capabilities it activates.Swipe the lifecycle

05 / 09

Stage 05 / Evidence · Verification

Does the exact candidate satisfy the specification?

Evaluate the produced candidate independently and bind every finding to the exact subject.

State earnedCandidate verified
Durable record
Evidence · Verification
Evidence required
Current, attributable evidence plus independent checks against each acceptance and policy criterion.
Factory coordination
Collect evidence, run independent checks, preserve lineage, and fail closed on missing proof.

Factory capabilities

9 of 10 participate at Verify
Active capabilities are marked in green
Participates at Verify

Evals

Test the exact candidate against acceptance, quality, security, and policy criteria.

Explore the Trust area

Why FDLC

Autonomy becomes useful when the surrounding system is trustworthy.

01

Intent stays explicit.

The factory begins with a governed outcome and acceptance contract.

02

Authority stays bounded.

Agents, models, and tools act inside declared operational limits.

03

Evidence earns progress.

Execution, verification, approval, release, and outcomes remain distinct.

Use FDLC now

Choose the next useful step.

Start with the depth that matches your question. Every path leads back to one governed system.

Understand

See one workflow move from intent to verified outcome.

Follow the five-minute example →Assess

Identify your current factory maturity and the next capability to earn.

Assess your factory →Build

Turn one engineering outcome into a reusable Factory Definition.

Build a Factory Definition →Read

Use the Guide as the field book for the deeper system.

Open the Guide →

Open-source reference implementation

See the operating model in Mission Control.

Inspect how plans, Work Orders, attempts, evidence, approvals, releases, and outcomes stay visible across autonomous software work.

Inspect Mission ControlView source ↗
MISSION CONTROL / COMMAND CENTERDEMO VIEW
Mission Control command center showing ranked work that needs operator attention
Deterministic demo view. Current capability status is documented on the Mission Control page.

What is real today

Framework, product, runtime, and experiments stay distinct.

These labels describe current public status. They are not certification and they do not invent availability.

  1. Framework

    FDLC

    Available

    Open operating model and engineering discipline.

  2. Book

    Guide

    Available

    The field guide that teaches how to implement FDLC.

  3. Runtime

    Mission Control

    Available

    Open-source control plane with capability-specific evidence and limits.

  4. Product

    AI-FDE

    Preview

    Internal alpha with a synthetic demo. Discovers and designs a Factory Definition; a Factory Deployed Engineer remains the accountable human.

  5. Executor

    Fab

    Experimental

    Local factory agent builder. No live model qualification yet.

  6. Platform

    Enterprise

    Proposed

    Commercial multi-factory control plane. Not generally available.

Need help applying FDLC?

Start with one bounded factory line.

AI-FDE helps an accountable human Factory Deployed Engineer turn reviewed enterprise evidence into an approved Factory Definition for Mission Control validation.

Explore AI-FDEEnterprise path →