BuiltForProd Blueprints

Build the solution

Production architectures for real workloads, deployed into the workload accounts of your Baseline. Each one is a working system, not a diagram.

What a Blueprint is

A BuiltForProd Blueprint is an opinionated, production-ready reference architecture for a real workload, delivered as repositories and deployed into the workload accounts of your BuiltForProd Baseline. It arrives as running infrastructure with its own pipelines, its own least-privilege deployer role, its own state and its own documentation — not as a diagram, a whitepaper or a starter template.

The distinction matters because the hard parts of a workload architecture are the ones a diagram leaves out: how a release is promoted, who can read which secret, what happens when a node dies, which findings block a deployment. A Blueprint answers those in code that is already running in your environment.

A working system, not a diagram.

How a Blueprint sits on the Baseline

It reads the foundation, it does not rebuild it

The Baseline owns the accounts, the address plan, the subnets, the hosted zones and the guardrails. A Blueprint reads those facts from Parameter Store at deploy time. There are no shadow VPCs, no duplicate identity, no second source of truth.

Its own role, its own state

Each Blueprint deploys through its own least-privilege deployer role and keeps its own state. The foundation and the workload move independently, so a workload change cannot reach into the landing zone.

Staged like the Baseline

The same stages the Baseline creates: sandbox, dev, staging and production. Staging and production are highly available; dev stays deliberately minimal, because paying production prices for a development environment is a decision, not a default.

Governed by the same pipeline

Every change is a pull request with a plan attached, reviewed under CODEOWNERS and applied by the pipeline. Production stays gated by required reviewers. The workload inherits the foundation’s change control rather than inventing its own.

The catalog

Three Blueprints today. More follow as the patterns that keep recurring get built.

Web Application Blueprint

A containerized web application on Amazon EKS, released through GitOps.

Stack
EKS with Karpenter, DocumentDB, ElastiCache for Redis, ALB behind WAF, CloudFront with a private S3 bucket, ArgoCD and Helm, External Secrets Operator.
What it is built around
Build once, promote the same immutable image from dev to staging to production.

Data & ETL Blueprint

A governed, event-driven data lake with no always-on compute.

Stack
Three-zone S3 data lake, AWS Glue catalog, PySpark job and crawler, a Lambda trigger, Lake Formation governance with LF-Tags and column-level grants.
What it is built around
Upload a JSON file under input/ and the pipeline runs end to end.

Secrets Blueprint

Secrets encrypted in Git, decrypted only by the identities that should.

Stack
SOPS with a KMS key per stage, synced to SSM Parameter Store as SecureStrings on merge, delivered to workloads by External Secrets Operator.
What it is built around
A developer with repository write access still cannot read a production secret.

How a Blueprint is delivered

  1. Scoped with your workload in mind

    We agree the stages, the sizing and the options before anything is deployed. A Blueprint is opinionated, but the opinions are visible and the deployment-specific values are tagged in the code.

  2. Repositories into your organization

    The Blueprint repositories are customized and created in your own version control organization, as forks of the upstream repositories so later improvements arrive as pull requests you review.

  3. Deployed by our engineers

    Into the workload accounts of your Baseline, stage by stage, through the same pipeline your team will use. Nothing is applied from a laptop.

  4. Handed over hands-on

    Your engineers perform the real operations themselves — deploy a change, promote a release, rotate a secret, read the logs — before sign-off. Then it is yours.

Blueprints come with a new Baseline as a Full Deployment with any two Blueprints, or later as one or more additional Blueprints, each individually priced. What shapes a quote.

Questions about Blueprints

What is a BuiltForProd Blueprint?

A BuiltForProd Blueprint is an opinionated, production-ready reference architecture for a real workload, delivered as repositories and deployed into the workload accounts of your BuiltForProd Baseline. It is a working system, not a diagram.

Do we need a Baseline first?

Yes. A Blueprint deploys into the workload accounts of a BuiltForProd Baseline and reads account and network facts from it, so the foundation has to exist first. You can buy both together as a Full Deployment with Blueprints, or add Blueprints later.

What the Baseline deploys

How is a Blueprint different from a Terraform module or a sample repo?

A module is a component and a sample is a demonstration. A Blueprint is a complete workload architecture, deployed and running in your own accounts by our engineers, with its own pipelines, its own least-privilege deployer role, its own state and its own documentation set.

Will a Blueprint conflict with our landing zone?

No. Blueprints read the facts they need — account identifiers, subnets, hosted zones — from Parameter Store, and never recreate what the landing zone owns. Each Blueprint has its own deployer role and its own state, so the foundation and the workload change independently.

Can we change a Blueprint after handover?

Yes, and most teams do. The repositories are yours, the code is source-available under the PolyForm Internal Use License 1.0.0, and everything ships with the golden-path guides your engineers need to add a service the same way.

How many Blueprints can we add?

As many as you want. Existing Baseline customers can add one or more additional Blueprints at any time, each individually priced and deployed into the workload accounts that already exist.

How engagements are priced

Blueprints assume a foundation: start with the Baseline. They are built to the same eight properties as everything else: read the Standard. If you want engineers to keep operating them with you, there is Managed.

Find out what production would take.

Tell us what you are running and what is coming: a funding round, an audit, a first enterprise customer, a migration. We will tell you what we would build, what it costs, and whether we are the right fit.