BoltPipeline logo

Environments

Every certified pipeline moves through Development, Integration, and Production. You cannot skip a step — and the same artifact runs in all three.

How it works

Three environments, one governed path

Every BoltPipeline tenant gets three environments: Development, Integration, and Production. A certified pipeline version must pass through them in order. Skipping Integration and deploying straight to Production is not permitted.

This is intentional. Each environment is a gate — a checkpoint where your team validates the pipeline before it reaches the next level. The Development validation is technical. The Integration validation is business. Production is the live system.

Dev

Development

Int

Integration

Prd

Production

Environment details

Dev

Development

Where your pipeline starts after certification. The agent assigned to your Development environment picks up the certified artifact on its next heartbeat and begins execution against your development data warehouse.

  • Validate that the SQL executes correctly against real (or representative) data
  • Smoke-test schema compatibility — catch column mismatches before they reach Integration
  • Iterate quickly: if something is wrong, create a new version, re-certify, and re-deploy to Dev

All changes require a new version and a new certification cycle. You cannot edit the SQL of an already-certified pipeline version.

Int

Integration

The staging gate before Production. Your pipeline runs against a near-production dataset. Use Integration to run quality checks, confirm data volumes, validate business rules with stakeholders, and get sign-off before any Production exposure.

  • Business validation with real-scale data
  • Cross-pipeline dependency testing
  • UAT sign-off by data consumers and business owners
  • Final performance and volume validation

Promoting to Integration requires at least Operator role.

Prd

Production

Live execution against your production data warehouse. The exact SQL that was certified is what runs in Production — no changes are possible. Drift monitoring is fully active and alerts trigger on any detected anomaly.

  • Live data pipeline execution on production data
  • Continuous drift monitoring (schema, volume, freshness, PII)
  • Full execution history and audit trail

Promoting to Production requires Admin role. Only one pipeline version can be active per environment at a time.

Core guarantee

Version immutability

Once a pipeline version is certified, its SQL is locked. The exact artifact that passed certification is what runs in every environment — Dev, Integration, and Production all execute the same code. There is no way to “hotfix” a pipeline in Production without going through a new certification cycle.

Why immutability matters

No production surprises

What passed certification in Dev is exactly what runs in Production. There is no drift between environments caused by manual edits.

Auditable history

Every version is timestamped and attributed. You always know who certified what, and when. Your compliance team will thank you.

Rollback is safe

Rolling back to a previous version means running a previously certified artifact — not reverting untested code changes.

Change control enforced

Any change to SQL requires a new version, new certification, and a fresh promotion cycle. No backdoors.

Rules

Promotion rules

Order is enforced

Dev → Integration → Production. You cannot promote a pipeline to Production until it has run in Development and Integration.

Role-based promotion

Promoting to Integration requires at least Operator role. Promoting to Production requires Admin role.

One active version per environment

Only one version of a pipeline can be active in each environment at a time. Promoting a new version automatically deactivates the previous one.

Rollback

You can roll back to the previously active version in any environment without a new certification cycle — the certified artifact already exists.

Need more environments?

Additional environments beyond Dev / Integration / Production are available on enterprise plans. Contact support@boltpipeline.ai to discuss your requirements.

Turn SQL into Production-Ready Data Pipelines — Faster and Safer

SQL-first pipelines, validated and governed — executed directly inside your database.

No new DSLs. No fragile orchestration. Just SQL with built-in validation, lineage, and governance.