Ce parcours n’est pas encore traduit en français : son contenu s’affiche en anglais.
DevOps Engineer
Automate, scale, and stabilize the infrastructure that powers engineering teams
Pas sûr(e) que ce soit pour vous ? Trouver ma voie →
0 missions sur 9 faites
Ce métier est-il fait pour vous ?
Deux minutes avant d’y consacrer des mois
Cochez ce qui vous ressemble
Cochez celles qui sont vraies pour vous.
Les compétences que vous en tirerez
Les emplois auxquels il mène
- DevOps Engineer
- Site Reliability Engineer (SRE)
- Platform Engineer
- Cloud Engineer
Avant de commencer
- ✓ Solid software engineering basics
- ✓ Familiarity with web application architectures
9 missions jusqu’à DevOps Engineer
4 mondes, 9 missions. Touchez un cercle pour l’ouvrir ; celui qui brille est le suivant.
DevOps starts with dominating the terminal, administering Linux, and automating tasks. You cannot automate what you cannot do manually.
Containers are the heart of modern software deployment. You need to know how to build them efficiently and orchestrate them at scale using Kubernetes.
Automate how code gets tested and released, and define your infrastructure purely as software code.
Architect systems in the cloud and learn how to monitor them effectively so you know when things break before the users do.
En savoir plus sur ce métier
Le métier, le salaire là où vous vivez, les questions fréquentes, et où ce métier peut vous mener ensuite.
Le métier
Ce que vous feriez
- Configure servers to meet functional specifications
- Monitor systems to catch problems early
- Install and upgrade software and hardware as ongoing maintenance
- Test and verify software patches and fixes
- Check that systems are stable, secure and able to scale
- Automate routine tasks and respond to alerts from automated monitoring
Où vous travailleriez
- Computer systems design firms
- Schools, colleges and universities
- Finance and insurance
- Information sector companies
- Company head offices
Toutes les étapes en détail
Le texte complet de chaque étape, à lire ou à imprimer. Avancez étape par étape dans le parcours ci-dessus.
Linux & Scripting Fundamentals 3 étapes
Master the Linux Terminal
Linux is the operating system of the cloud. You must move past the GUI and become extremely comfortable navigating, searching, and managing files entirely from the command line.
Tâche concrète: Spin up a free Ubuntu VM on AWS (EC2) or DigitalOcean. SSH into it. Find all `.log` files in `/var/log` modified in the last 7 days, extract all lines containing the word 'error' (case-insensitive), and append them to a new file in your home directory using a single chained command (piping).
Pourquoi c’est important: When a production server goes down at 2 AM, there is no UI. You only have SSH and the terminal to diagnose why memory is maxed out or why a service crashed.
Liste de contrôle
- Navigate the filesystem fluidly (cd, ls, pwd, find, locate)
- Master file manipulation and text processing (cat, grep, awk, sed, tar, gzip)
- Understand standard streams (stdin, stdout, stderr) and piping/redirection
- Manage users, groups, and permissions (chmod, chown, visudo)
- Monitor system performance and processes (top, htop, ps, kill, free, df)
Networking & Security Basics
Servers must talk to each other securely. You must understand how traffic flows across the internet and within private networks.
Tâche concrète: Install Nginx on your Linux VM. Configure a reverse proxy that listens on port 80 and forwards traffic to a dummy python application running on port 8080. Securing it: configure UFW (firewall) to block all incoming traffic EXCEPT ports 22 (SSH) and 80 (HTTP).
Pourquoi c’est important: Fixing 'the app can't reach the database' is your job. If you don't understand subnets, routing, and DNS, you will be guessing instead of troubleshooting.
Liste de contrôle
- Understand the OSI Model and TCP/IP stack
- Master DNS fundamentals (A records, CNAME, TXT, how resolution works)
- Learn how to troubleshoot connectivity using ping, traceroute, netstat/ss, and dig/nslookup
- Understand firewall basics (ufw, iptables) and SSH key-based authentication
- Grasp the concepts of Load Balancing, Reverse Proxies (Nginx), and HTTPS/SSL certificates
Scripting (Bash & Python)
You need to write code that automates repetitive system tasks. Bash is for tying tools together; Python is for complex logic.
Tâche concrète: Write a Bash script that performs a backup. It should compress a target directory into a `.tar.gz` file, append the current timestamp to the filename, move it to a `/backups` directory, and automatically delete any backups older than 7 days purely through shell commands. Schedule it to run daily using `cron`.
Pourquoi c’est important: Manually SSHing into 50 servers to update a config file takes all day. A Bash script does it in 5 seconds. Automation is the 'Auto' in DevOps.
Liste de contrôle
- Write Bash scripts utilizing variables, loops, conditionals, and functions
- Handle command-line arguments and exit codes in Bash
- Learn Python basics specifically geared toward system administration
- Write Python scripts that interact with APIs (using the `requests` library)
- Use Git for version control of all your scripts
Containers & Orchestration 2 étapes
Containerization with Docker
Docker packages code and dependencies so an app runs the exact same way everywhere. It solves the 'it works on my machine' problem.
Tâche concrète: Take a simple Node.js or Python API. Write a multi-stage Dockerfile that compiles/installs dependencies in stage 1, and only copies the built artifacts to a distroless runtime image in stage 2. Your final image size should be under 150MB. Verify it runs via `docker run`.
Pourquoi c’est important: Bloated Docker images slow down deployment pipelines and increase security vulnerabilities. Creating lean, secure containers is a baseline DevOps skill.
Liste de contrôle
- Write highly optimized Dockerfiles (multi-stage builds, minimizing layers, using Alpine images)
- Manage container lifecycles (run, stop, rm, exec, logs)
- Understand Docker networking (bridge, host, overlay networks)
- Manage persistent storage using Docker Volumes
- Use Docker Compose to declare and run multi-container applications (e.g., App + Database + Cache)
Orchestration with Kubernetes (K8s)
When you have 500 containers running across 20 servers, Docker Compose isn't enough. Kubernetes automatically heals, scales, and orchestrates containers.
Tâche concrète: Install Minikube or kind locally. Write YAML manifests to deploy a web application and a Redis cache. Configure a Deployment with 3 replicas for the web app. Create a Service so the web app can talk to Redis. Cause chaos by manually deleting one of the web app Pods and use `kubectl get pods -w` to watch K8s instantly spin up a replacement.
Pourquoi c’est important: Kubernetes is the modern operating system of the cloud. Almost all large tech companies use it to manage their microservices architecture.
Liste de contrôle
- Understand the K8s Architecture (Control Plane, Kubelet, Worker Nodes)
- Master Core Objects: Pods, ReplicaSets, Deployments, and DaemonSets
- Master Networking Objects: Services (ClusterIP, NodePort, LoadBalancer) and Ingress
- Learn ConfigMaps and Secrets for managing environment variables
- Understand persistent storage (PV, PVC, StorageClasses)
- Learn to use `kubectl` fluidly
CI/CD & Infrastructure as Code (IaC) 2 étapes
Continuous Integration & Delivery (CI/CD)
Stop deploying from developers' laptops. Build pipelines that automatically test code and push it to production.
Tâche concrète: Create a GitHub Repository for a web app. Write a GitHub Actions Workflow (`.github/workflows/deploy.yml`) that triggers when code is pushed to `main`. It should: 1. Run unit tests. 2. If tests pass, Build a Docker image. 3. Log in to DockerHub and Push the image. 4. SSH into your VPS (using GitHub Secrets) and run a script to pull and restart the container.
Pourquoi c’est important: Manual deployments cause human error and downtime. A robust CI/CD pipeline enables teams to deploy safely 10+ times a day.
Liste de contrôle
- Understand CI (automated testing/linting) vs CD (automated deployment)
- Familiarize yourself with GitHub Actions or GitLab CI syntax
- Write pipelines that trigger on Pull Requests
- Handle secrets securely within CI tools (e.g., AWS credentials)
- Learn how to push Docker images to registries (DockerHub, ECR) inside pipelines
Infrastructure as Code (Terraform)
Clicking buttons in the AWS console is prone to error and isn't version-controlled. Define your servers in code.
Tâche concrète: Sign up for AWS (Free Tier). Write a Terraform script that automatically provisions a VPC, public/private subnets, Security Groups, and an EC2 instance running Nginx. Run `terraform apply` to create the whole architecture in 2 minutes. Then run `terraform destroy` to tear it all down cleanly.
Pourquoi c’est important: IaC allows you to spin up an exact replica of your production environment for staging in minutes, and tracks changes to infrastructure in Git just like application code.
Liste de contrôle
- Understand the declarative nature of Terraform
- Master Terraform state tracking (`terraform.tfstate`) and why remote state is critical
- Write HCL (HashiCorp Configuration Language) for variables, providers, and resources
- Use Terraform Modules to keep configurations DRY (Don't Repeat Yourself)
- Understand `terraform plan` vs `terraform apply`
Cloud Providers & Monitoring 2 étapes
Cloud Architecture (AWS/Azure/GCP)
Grasp the core services of a major cloud provider. AWS is the industry standard.
Tâche concrète: Architect a highly available 3-tier application on AWS natively (without Terraform for learning). Create a VPC spanning two Availability Zones. Set up an Application Load Balancer that routes traffic to an Auto Scaling Group of EC2 instances, hooked up to an RDS Multi-AZ database.
Pourquoi c’est important: Companies migrate to the cloud for reliability and elasticity. If you don't know how to architect across Availability Zones or manage IAM roles, you will build systems that are fragile and insecure.
Liste de contrôle
- Master Identity and Access Management (IAM): Users, Roles, Policies, Principle of Least Privilege
- Understand Compute: EC2, Auto Scaling Groups (ASG), and Serverless (Lambda)
- Understand Storage: S3 (Object storage), EBS (Block storage)
- Understand Databases: RDS (Relational) vs DynamoDB (NoSQL)
- Familiarize yourself with Cloud Networking: VPCs, NAT Gateways, Route53 (DNS)
Observability & SRE Principles
Logging and monitoring are your eyes and ears in a distributed system.
Tâche concrète: Deploy Prometheus and Grafana onto your Linux VM/K8s cluster. Configure Prometheus to scrape CPU and Memory usage metrics from Node Exporter. Build a Grafana dashboard visualizing that data. Set an alert rule so that if CPU usage spikes above 80% for 5 minutes, it fires an alert to a Discord or Slack webhook.
Pourquoi c’est important: Without monitoring, you only find out the site is down when a customer complains on Twitter. Proactive alerting solves issues before they escalate into outages.
Liste de contrôle
- Understand the Three Pillars of Observability: Metrics, Logs, and Traces
- Set up Prometheus for scraping metrics and Grafana for visualizing dashboards
- Implement centralized logging using ELK (Elasticsearch, Logstash, Kibana) or Promtail/Loki
- Understand SRE concepts: SLIs (Service Level Indicators), SLOs, and Error Budgets
- Configure automated alerting (PagerDuty, Slack integration) based on metric thresholds
Questions fréquentes : DevOps Engineer
Des réponses claires sur la durée, les prérequis, les salaires et les méthodes d’étude.
Combien de temps faut-il pour devenir DevOps Engineer ?
Ce parcours estime 6–12 months. Il compte 4 phases et 9 étapes, de « Linux & Scripting Fundamentals » à « Cloud Providers & Monitoring ».
Comment devenir DevOps Engineer ?
Suivez les 4 phases dans l’ordre, en commençant par la phase 1, « Linux & Scripting Fundamentals ». Votre première tâche : Spin up a free Ubuntu VM on AWS (EC2) or DigitalOcean. SSH into it. Find all `.log` files in `/var/log` modified in the last 7 days, extract all lines containing the word 'error' (case-insensitive), and append them to…
Combien gagne un DevOps Engineer ?
Le salaire dépend du pays. L’encadré salaire de cette page indique l’estimation de DraftMyCareer, en monnaie locale, pour le pays choisi dans le menu.
De quoi ai-je besoin avant de commencer ?
Ce parcours indique : Solid software engineering basics; Familiarity with web application architectures.
De quelles compétences un DevOps Engineer a-t-il besoin ?
Ce parcours couvre : Linux, Docker, Kubernetes, CI/CD, AWS / Azure, Terraform, Python / Bash.
Changer de métier : passerelles vers ce poste et depuis ce poste
Voir toutes les reconversions →Évolutions et changements de poste possibles
- Cloud Architect
Cloud Architects define strategic technical direction for enterprises, earning top-tier staff compensation.
⏱ 4–6 months🧭 ModerateCe que vous apprendriezMulti-Region High Availability · Enterprise Well-Architected Framework · Cloud Cost Optimization (FinOps)
Voir le plan d’action
- Month 1–2: Cloud Architecture Certification Study and obtain AWS Solutions Architect Professional or Google Cloud Architect.
- Month 3–4: Enterprise Distributed Systems Design resilient multi-AZ architectures, event-driven pipelines, and VPC peering topologies.
- Month 5–6: FinOps & Governance Implement organization-wide cost optimization, security boundaries, and SLA governance.
Venir à ce poste depuis un autre métier
- Depuis: Backend Developer
DevOps & Platform engineers are among the most sought-after tech specialists with high job security and remote leverage.
⏱ 4–6 months🧭 ModerateCe que vous gardezLinux CLI · Networking basics · Server runtime knowledge
Voir le plan d’action
- Month 1: Advanced Linux & Networking Master Linux kernel diagnostics, systemd, TCP/IP, DNS, and shell scripting.
- Month 2: Containers & Orchestration Containerize complex apps with multi-stage Docker builds and manage K8s deployments.
- Month 3: Infrastructure as Code Provision multi-environment cloud infrastructure using Terraform and AWS/GCP.
- Month 4–6: Production SRE & Observability Build automated CI/CD release trains, zero-downtime rollouts, and alert metrics.