I have about eight years of IT experience at a small-to-medium business MSP, including Intune administration for roughly 15,000 devices, project engineering, systems administration, Veeam, basic networking, helpdesk, and resolving compliance issues found during audits.
A friend is helping connect me with interviews for an apprentice DevOps position. The team works with Pulumi, GitOps, AWS, GCP, and Kubernetes. I have spent around six months building a homelab and experimenting with these types of tools, often using AI to help me find and understand documentation.
The initial project would involve improving SOC 2 compliance in the infrastructure by modifying Pulumi code written in Go. I do not currently have practical Go experience, and I have not worked in a production DevOps environment before. However, the company knows that I am coming in to learn, and the longer-term plan would be to develop my skills and eventually move me into a junior engineering role.
If senior engineers review my changes before they reach production, is this a reasonable transition to make? Could someone with my background become productive in three to six months, or am I taking on too much at once?
5 Answers
The biggest transition is usually not learning one particular tool. It is understanding how developers build and release software, plus learning to create self-service systems instead of manually handling every request. Books such as The Phoenix Project, Accelerate, and The DevOps Handbook can help you understand the practices behind the technology. The tools can be learned as you apply those concepts.
This sounds challenging, but not unreasonable if the apprenticeship is genuinely structured around learning. Your infrastructure, troubleshooting, networking, backup, and compliance experience all transfer well. Go syntax is probably not the biggest concern; the more important issue is having experienced engineers review your work before it affects production. Three to six months may be enough to become useful, but not enough to master everything.
There are valid reasons to be cautious. Production DevOps work often requires strong Linux, networking, cloud, security, and programming fundamentals, and learning all of that at once can be overwhelming. If the company expects you to independently design infrastructure from day one, this may be too much. If they expect you to learn while working on existing configurations with careful reviews, it is a very different and much more manageable situation.
Start with the parts closest to your current experience: AWS and GCP fundamentals, Pulumi, IAM, networking, logging, and compliance controls. Kubernetes can be a much bigger jump, especially if your Linux experience is limited, so it may make sense to tackle that later. Before writing new infrastructure, spend time reading the team's existing Pulumi code and learning how they structure, test, and review changes.
Do not let the lack of Go experience automatically stop you. If you can read code, learn consistently, and are willing to investigate problems instead of guessing, you can pick up enough Go to work with Pulumi. The SOC 2 portion may actually require more domain learning than the language itself, since you will need to understand the controls, evidence, access policies, and operational processes behind the infrastructure changes.
That is reassuring. I should have several senior engineers reviewing changes, and a technical lead has offered to help train me and answer questions while I ramp up.