The Infrastructure Factory

The definition, taken apart

An Infrastructure Factory is a platform¹ that turns intents², from humans or agents³, into compliant, owned, observable, cost-attributed⁴ environments⁵, with no ticket in the path⁶.

1. 1

   Platform

   A production system that a team runs as a product, with customers, a backlog and a service level. Not a project, not a folder of modules, not a committee of approvers. Its operators build the line.

2. 2

   Intents

   A request that says what the environment is for, who owns it, what data it touches and how long it lives. The factory translates the intent into resources. The requester never has to.

3. 3

   Humans or agents

   The same contract serves a developer in a portal, a delivery pipeline, an operations agent remediating an incident, and a coding agent validating a change in the middle of the night. If an agent cannot use it unaided, it is a portal, not a factory.

4. 4

   Compliant, owned, observable, cost-attributed

   Four properties present on arrival, never added later. Compliant: every policy evaluated, every decision recorded. Owned: a cost center, a compliance owner and an on-call owner that resolve to real people. Observable: telemetry flowing before the first request arrives. Cost-attributed: every euro lands on the owner's line.

5. 5

   Environments

   The factory's unit of product: a bounded set of resources with a name, an owner, a lifetime, a data classification, a network position, an identity, telemetry and a cost line. An account, a namespace, a cluster, a sandbox or a set of managed services can each be an environment. A virtual machine is a part.

6. 6

   No ticket in the path

   No human action between the intent and the environment. Humans decide at the contract, once, and the decisions live on as defaults, delegations and policies. A human may be notified, asked to approve a renewal above a threshold, or paged when a policy fails. A human is never a station.

What it is made of

Each of the parts exists as open source or as a cloud primitive today. The factory is the arrangement, not any one of them.

What it is not

The table reads left to right as a history: most organizations have built the first three and call the result a platform.

It is also not an AI factory in NVIDIA's sense, not an IaC generator, and not a product from any one vendor. A portal is its front door. An orchestrator is one of its stations. An IaC generator is a faster way to write the input the factory exists to make unnecessary.

From ticket desk to factory

Most organizations are somewhere on this ladder. Each stage keeps everything the previous one built and removes one kind of human from the path. The test at every stage is the same: can an agent with a scoped identity obtain an owned, compliant environment at any time?

1. 0

   Ticket desk

   Requests arrive as text. People build what they understand. Lead time is days to weeks, and the organization measures the queue.

   Fails the test: nobody is awake.

2. 1

   Automated fulfilment

   Scripts and modules do the building, but a person still approves, classifies and assigns cost. Lead time drops; the queue stays.

   Fails the test: the approval is a person.

3. 2

   Self-service portal

   Humans fill forms. Decisions are still made per request, behind the form, and the form itself is unreadable to an agent.

   Fails the test: the agent cannot fill the form, and the form cannot sign the owner.

4. 3

   Contracted API

   Machine-readable contracts, a policy pre-check with readable denials, defaults for ownership. Agents can ask. Evidence, lifetime and capacity are still somebody else's problem.

   Fails the test at volume: orphans accumulate and the audit is compiled by hand.

5. 4

   Infrastructure Factory

   Ownership by default, identity for every requester, a lifetime on everything, evidence shipped with the product, capacity held as inventory, and the factory publishing its own metrics.

   Passes the test, at production volume, on every terrain.

You have an Infrastructure Factory when

• An agent with a scoped identity can obtain an owned, compliant environment with nobody awake.
• Every environment has a reachable owner, a cost center and a classification before it exists.
• Every denial carries a reason the requester can act on without opening a ticket.
• Every environment has a lifetime, and reclaim runs without a human.
• Every environment ships with its decision record.
• The factory publishes its own lead time, zero-touch rate, time to owner and orphan rate.
• A contract can move to another terrain without changing what the requester asks for.
• Platform engineers spend their time on contracts, policies and lines, and none of it fulfilling requests.

1. 1968

   The term is coined

   Bob Bemer proposes the "software factory".

2. 1975

   The first one runs

   System Development Corporation operates its Software Factory.

3. 2004

   The book

   Jack Greenfield and Keith Short publish Software Factories with Wiley and Microsoft.

4. 2017

   Kessel Run

   The US Air Force launches "DOD's first software factory".

5. Dec 2019

   Platform One

   Established to "assist software factories by helping them focus on building mission applications": DevSecOps as a service, Kubernetes, Iron Bank hardened images and the Big Bang tooling.

   It was called a software factory. It was an infrastructure platform.

6. 2021

   The lesson, stated

   Air Force CIO Lauren Knausenberger: "starting with Kessel Run smuggling DevSecOps into the DOD, and continuing with Platform One leading the way to Kubernetes, a common repo, and a desire to bring the entire community together and leverage common enterprise services."

7. 2025–26

   The agentic software factory

   Factory.ai's "Factory 2.0" (June 2026), BCG Platinion's "Agentic Software Factory" (March 2026), StrongDM's no-human rules (February 2026) and Uber's "Running a Software Factory Efficiently at Uber Scale" (August 2026). In Factory.ai's words, "the incremental units of this system are AI agents."

8. 2026 →

   The Infrastructure Factory

   The platform the agentic software factory runs on, named as a production system of its own, with its own line and its own metrics.

1. Intent arrives

   The agent calls a tool exposed over MCP or a plain API: an environment for the service orders-api, with Postgres, a public endpoint, classification derived from its data sources, for 48 hours. A human in a portal sends the same request through a form. Nobody writes Terraform here.

   What moves to the next station

   A structured intent, as YAML or JSON, validated against a published schema.

   Building blocks

   • MCP tool server
   • Backstage
   • Portal or CLI

2. Golden-path contract

   The intent resolves to a published abstraction: a Kratix Promise or a Crossplane Composition. The contract is versioned and machine-readable. It lists what the requester may choose and what the platform decides, which is exactly the information an agent needs to ask for something fulfillable the first time.

   What moves to the next station

   A claim against a named contract version.

   Building blocks

   • Crossplane
   • Kratix
   • Backstage software templates

3. A policy pre-check the agent can read

   OPA or Kyverno evaluates the claim before anything is created and returns allow or deny with a reason the agent can act on. Defaults fill the fields the agent cannot sign: the cost center from the owning team's catalog record, the compliance owner by delegation, the classification from the declared data. A denial is a usable answer, and never a ticket.

   What moves to the next station

   An admitted claim with ownership, classification and budget attached, plus the policy decision kept as evidence.

   Building blocks

   • Open Policy Agent
   • Kyverno
   • Gatekeeper

4. Vend

   The vending lines that already exist do their work: an account or project from AWS Account Factory for Terraform, Azure subscription vending or Google's project factory; a cluster or namespace from Cluster API or a virtual cluster; the Postgres and the endpoint from OpenTofu modules. GitOps reconciles the result and keeps reconciling it.

   What moves to the next station

   A reconciled environment, with endpoints and a record of what was created.

   Building blocks

   • OpenTofu
   • Argo CD
   • Flux
   • Cluster API (Kubernetes SIG)
   • Account Factory for Terraform

5. Execution sandbox for untrusted code

   When the environment exists to run code an LLM wrote, a Kubernetes Agent Sandbox supplies the isolated runtime from a warm pool, claimed in seconds. The sandbox is one station and never the whole factory. Stations three, six and seven are what the CNCF means by "sandboxing your agent is not enough".

   What moves to the next station

   A SandboxClaim bound to a running, isolated workload.

   Building blocks

   • Agent Sandbox (SIG Apps)
   • gVisor or Kata Containers

6. Attach what humans used to add by hand

   Workload identity for the environment, and a scoped, least-privilege identity for the agent that asked for it. Telemetry wired to the platform's collectors. Cost attribution tags that match the cost center from station three. The ownership record registered in the catalog. None of this needs a person, and all of it used to wait for one.

   What moves to the next station

   An environment that is observable, attributable, and reachable only by the identities the policy allowed.

   Building blocks

   • SPIFFE / SPIRE
   • OpenTelemetry
   • OpenCost
   • Backstage catalog

7. Hand over, then reclaim

   The requester receives endpoints, credentials through the identity system, and the evidence bundle: policy decisions, owner, classification, budget. The TTL and the owner record make reclamation automatic. The factory emits its own metrics at this station: lead time, zero-touch, time to owner.

   What leaves the line

   An environment in use, a metrics record, and a scheduled reclaim.

   Building blocks

   • Prometheus
   • Kyverno cleanup policies
   • Argo CD or Flux

CNCF project    Other open source, Kubernetes SIG or cloud-provider building block

Cloud-likeness, defined

NIST's 2011 definition of cloud computing lists five essential characteristics: on-demand self-service, broad network access, resource pooling, rapid elasticity and measured service. An Infrastructure Factory consumes all five. A hyperscaler supplies them as APIs. Wherever one is missing, the factory has to manufacture it before it can vend anything, and that is the whole difference between the cloud factory and every other kind.

Two things NIST does not list but a hyperscaler also supplies: one identity system across every resource, and a certified compliance posture under a shared-responsibility model. Off the cloud, the factory owns both. Keycloak and SPIFFE/SPIRE take the first; the evidence bundle from station seven takes the second.

What changes on the line

Different terrains, require one contract

1. I

   Intent in

   The input is an intent: what the environment is for, who owns it, what data it touches, how long it lives. Never a resource list. A resource list is the factory's output, translated back into its inputs by a human.

2. II

   Environment out

   The output is an environment that is owned, compliant, observable and cost-attributed on arrival. Anything less is a resource, and resources are what the ticket queue used to deliver.

3. III

   Machine-readable contracts

   What the requester may choose and what the platform decides is a versioned schema. The API is the interface; portals and CLIs are views on it. Agents are first-class customers, and a contract an agent cannot read is a contract a human will be asked to interpret.

4. IV

   No ticket in the path

   Humans decide once, at the contract, or encode the decision as a default, a delegation or a policy. A human may be notified. A human is never a blocking station. Platform engineers build the line; they do not fulfil requests.

5. V

   Policy before provisioning

   The pre-check returns allow or deny, with a reason the requester can act on, before anything is created. Nothing is built to be torn down by a reviewer later, and a denial is an answer rather than an escalation.

6. VI

   Ownership by default

   Cost center, compliance owner, on-call owner and data classification are inherited from the catalog, never typed into a form. No reachable owner, no environment.

7. VII

   Scoped identity for every requester

   Every request runs under a short-lived, least-privilege identity, agents included. Credentials move through the identity system and never through a prompt, a ticket or a chat message.

8. VIII

   Everything expires

   A lifetime is part of every intent. Renewal is a request like any other and carries the same owner. Reclaim is automatic, and an orphan is a defect of the factory, not of the requester.

9. IX

   Evidence ships with the product

   The decision record travels with the environment: policies evaluated, owner, classification, budget, jurisdiction. An audit reads the factory's output, not a spreadsheet assembled afterwards.

10. X

    Capacity is inventory

    Warm pools, quotas and buffer stock absorb bursts. Where elasticity is missing, the factory keeps stock and publishes its fill rate instead of pretending that lead time is zero.

11. XI

    The factory measures itself

    Lead time to a compliant environment, environments per platform engineer, zero-touch rate, policy-violation rate, time to owner, orphan rate. Published to the factory's customers the way an SLO is, and argued about the same way.

12. XII

    Built from replaceable parts

    Contracts outlive tools. Every station is a building block that can be swapped, Crossplane for Kratix, Kyverno for OPA, a hyperscaler for a rack, without changing what the requester asks for or what the requester gets back.