"I know a guy who can fix that."
At Level 1, there is no formal platform. Infrastructure and deployments exist, but they are the product of individual heroics rather than shared systems. Think of a small restaurant where one chef knows every recipe from memory — the food is good, but if that chef calls in sick, the kitchen shuts down.
Characteristics:
- Process: Manual. Deployments happen via CLI commands from a developer's laptop, brittle shell scripts checked into no one's repository, or SSH sessions that nobody documents.
- Standardization: Non-existent. Team A uses AWS, Team B uses Azure. Team A uses Jenkins, Team B uses GitHub Actions. Team C uses a custom bash script that only works on macOS Monterey.
- Experience: "Ticket-driven" or "Hero-driven." If you need a database, you ask the one person who knows how to set it up. That person is always busy. You wait.
- Observability: Minimal. Logs exist on individual servers. Alerts are "the CEO calls because the site is down."
- Risk: Extremely high. The Bus Factor — the number of people who need to be hit by a bus before the system collapses — is often 1.
What failure looks like at Level 1: A critical production deployment fails at 2am. The only person who knows the deployment process is on vacation in a timezone with no cell service. The team spends four hours reverse-engineering a bash script with no comments.
Goal: Move from Hero-based to Team-based operations. The first win is documentation — writing down what the heroes know so that knowledge lives in the organization, not in individuals.