Next Steps
If you completed the Quick Start, you have a local Infrahub instance with a sample schema and some data. This page outlines the key steps to turn that into a proof of concept that demonstrates value to your organization.
1. Design your schema
The Quick Start used pre-built schemas from the Schema Library. For a POC, model a slice of your actual infrastructure — the devices, services, or resources that matter most to your team.
Start with a focused scope. Pick one domain (for example: data center fabric, WAN sites, or firewall policies) and model it. You can always expand later. The Schema Library remains a good starting point to extend from.
Create a schema../guides/create-schema2. Load real data
With a schema that reflects your infrastructure, bring in real data. If you already have a source of truth (Netbox, Nautobot, IP Fabric, or others), Infrahub Sync can pull data from these systems directly.
Alternatively, describe your data in YAML object files and load them with infrahubctl, the same way you did in the Quick Start. This works well for initial seeding or for data that doesn't live in another system yet.
3. Connect a Git repository
In the Quick Start, schemas and data files lived on your local machine. For a POC, store them in a Git repository and connect it to Infrahub. This unlocks Git-native workflows: Infrahub will sync branches, track changes, and consume files like Jinja2 templates and Python scripts directly from the repository.
Push the project created during the Quick Start to a remote repository (GitHub, GitLab, Bitbucket, etc.) and connect it through the UI or CLI.
Connect a repository to Infrahub../guides/repository4. Generate artifacts
Once you have real data and a connected repository, use Transformations to generate outputs from your data — device configurations, cabling plans, documentation, compliance reports, or anything else your team needs.
This is where Infrahub starts delivering tangible value: a single data change can regenerate all affected artifacts automatically.
Generate artifacts../guides/artifact5. Deploy to a shared environment
A local Docker setup works for development, but for a POC that involves your team, deploy Infrahub to a shared environment. Infrahub is packaged as Docker images and supports multiple deployment options:
- Docker Compose on a shared VM for quick team access
- Kubernetes via Helm Chart for production-grade deployments
Going further
At this point, you have a running Infrahub instance with your own schema, real data, a connected Git repository, and generated artifacts. From here you can:
- Build Generators to automate the creation of infrastructure objects from templates and business logic.
- Set up events and webhooks to trigger external systems when data changes.
- Deploy artifacts with Ansible or Nornir to push configurations to your infrastructure.
If you need guidance on which feature to explore next or how to implement a specific use case, reach out to the community on Discord or book a meeting with OpsMill.
Book a meeting with OpsMillhttps://cal.com/team/opsmill/meet