InfoIMAGE, Inc.At InfoIMAGE, our infrastructure team relied on manual VM provisioning through vSphere, a process that took hours per server and, across our fleet of 196 servers, led to significant configuration drift. This drift caused recurring outages and slowed down new deployments at a time when the company was actively scaling, making it clear that our provisioning approach could no longer keep pace with the business.
I led the design and build of a fully automated provisioning pipeline, creating reusable Terraform modules to handle VM creation and Ansible playbooks to standardize configuration. Rather than attempting a risky big-bang cutover, I worked with the infra team to migrate servers in batches organized by service tier, validating each batch thoroughly in staging before promoting it to production. This incremental approach let us catch issues early and avoid downtime during the transition.
The results were substantial: configuration drift incidents dropped to near zero, and outages caused by VM misconfigurations fell by more than 80%. Beyond the reliability gains, the automation freed up roughly 20 hours per week that the infra team had previously spent on manual provisioning, allowing them to redirect that time toward higher-value engineering work instead of repetitive server setup.
