Python dev (Django/FastAPI/Docker/K8s) trying to break into DevOps — what should I prioritize, and what are the real problems no one warns you about?

Аналитика рынка

Стек

Откликнуться

Загрузите резюме — мы свяжем вас с работодателем напрямую через нашу базу.

Описание

Python dev (Django/FastAPI/Docker/K8s) trying to break into DevOps — what should I prioritize, and what are the real problems no one warns you about? Hey everyone, long-time lurker, first time posting here. Looking for honest advice from people who've actually made this kind of transition. My current stack: Python · Django / FastAPI · Docker + Compose · Kubernetes (basics) · Redis / PostgreSQL · Celery / Async · Bash / Linux · RTSP / FFmpeg pipelines / LLMs · YOLO / OpenCV I've been building backend systems and a full AI-powered camera security system from the ground up — ingestion pipelines, async workers, containerized deployments, the whole thing. So I'm not starting from scratch, but I know my infra/ops knowledge has real gaps. Now I want to go deeper into the operations side — CI/CD pipelines, infrastructure-as-code, monitoring, cloud, reliability engineering. Basically bridge the gap between "I can Dockerize things" and "I own the entire deployment lifecycle." What I want to learn next: CI/CD pipelines end-to-end (GitHub Actions, GitLab CI, Jenkins?) Terraform or Pulumi for infrastructure-as-code Proper Kubernetes beyond just "kubectl apply" — RBAC, Helm, Ingress, autoscaling Cloud fundamentals — AWS or GCP (which is better to start with?) Observability stack — Prometheus, Grafana, ELK, alerting GitOps workflows — ArgoCD, FluxCD Real questions for this community: 1. What order should I learn these in? I've seen conflicting roadmaps. Some say start with cloud, others say master Linux first, others say just go build something and learn as you go. 2. What are the actual painful problems nobody tells you about? Not the beginner stuff — I mean the things that trip up even experienced engineers. The stuff that takes months to unlearn or figure out on your own. 3. Career reality check — I'm coming from a Python/ML background. Will that help me in DevOps roles or will recruiters just not take me seriously because I don't have a traditional sysadmin / infra background? The real problems I'm already anticipating (want your take on these): Tool sprawl confusion — Terraform vs Pulumi vs CDK vs Ansible vs Chef — no one agrees and every job posting wants something different. How did you pick one and stick with it? Cloud costs — I have zero experience budgeting cloud infra and I know this bites everyone at some point. Any war stories? Debugging distributed failures — logs scattered across 10 services, no clear owner, alerts firing at midnight. How long did it take you to get good at this? Kubernetes complexity cliff — goes from "simple" to genuinely hard very fast, and tutorials always skip the hard parts. What resource actually helped you get past that wall? "DevOps is a culture, not a role" — some companies don't even have a DevOps team, it's just dumped on top of dev work with no extra support or title. How common is this really? Imposter syndrome — coming in as a developer, not a sysadmin, means constantly feeling like you're missing some foundational Linux/networking knowledge everyone else just has. Did this get better? [link] [handle]

Контакты работодателя (email/phone/telegram) скрыты из публичного превью — отправьте резюме, чтобы мы связали вас напрямую.