Weekend Site Reliability Engineer

What you’ll be doing

• Work with a team of DevOps and DBA professionals; covering Saturday, Sunday and Monday (5 days in total with flexibility in your days off) as a Weekend SRE
• Improve existing infrastructure and processes across the countries we’re deployed in, as well as streamlining processes to deploy to new countries in the future
• Continuously improve Kubernetes platform stability and efficiency, with a focus on optimising resource utilisation, reducing costs, and streamlining environment provisioning through GitOps-first practices
• Monitor and maintain cloud infrastructure through autoscaling, alerting pipelines, and Grafana dashboards covering metrics, logs, traces, and real user monitoring (RUM)
• Own weekend on-call operations, triaging and responding to production incidents, performing root cause analysis, and driving post-incident reviews
• Design and manage alert pipelines to ensure actionable signal quality, with attention to preventing alert fatigue, waterfall alerting, and notification flooding
• Define and maintain SLIs and SLOs for critical services, and use them to drive reliability improvements and on-call prioritisation
• Take ownership and responsibility for our cloud operation activities
• Liaise with external security agencies for annual audits as well as perform our own internal security sweeps
• Aid in reconfiguring existing architecture to allow for rapid deployments to new countries
• Mentoring less experienced team members

What you’ll bring

• 3+ years DevOps / platform engineering experience
• Must be based in Europe or Asia or LatAM
• Experience independently leading the planning and deployment of a project
• Experienced with cloud platforms, especially AWS, including solid knowledge of how to utilise cloud resources to fulfil the demand from other teams and production
• Strong understanding of Kubernetes and container orchestration, with experience in EKS and GitOps tooling such as ArgoCD and Helm being highly valued
• Experience with Infrastructure-as-Code, particularly Terraform
• Proficiency in scripting and automation with Bash, Python, or Golang; experience with Rust is a plus
• Hands-on experience with observability stacks covering metrics, logs, distributed traces, and profiling, for example Prometheus, Loki, Tempo, Pyroscope, and OpenTelemetry
• Experience with real user monitoring (RUM), with familiarity in Grafana Faro or OpenTelemetry SDK instrumentation being a plus
• Proven on-call and incident response experience, comfortable triaging production issues under pressure, leading post-mortems, and driving follow-up actions
• Ability to design and maintain alert frameworks that minimise noise, prevent alert fatigue, and avoid waterfall alerting patterns
• Experience defining SLIs and SLOs and using them to inform reliability work
• Familiarity with service mesh concepts is a plus, as we are actively evaluating Cilium-based service mesh in non-production environments
• Solid networking knowledge, especially the TCP / IP stack and HTTP protocol
• Experience handling high HTTP request volumes and designing systems for high availability and high traffic environments
• A strong understanding of cache, including CDN, HTTP cache, Redis / Memcached
• Excellent troubleshooting skills, including Linux OS issue diagnosis and OS parameter optimisation, JVM optimisation would be highly advantageous

Our stack

• Languages: Java / Spring Boot, Node.js, Python, JavaScript
• Database: Aurora MySQL & PostgreSQL, MongoDB, MySQL Community
• Cache: ElastiCache, Redis, Valkey
• Messaging: Apache RocketMQ, AutoMQ, Kafka
• Networking & Proxy: Nginx, Kong, Cilium, eBPF
• Orchestration & GitOps: Docker, Kubernetes (EKS), ArgoCD, Helm
• Computing & Storage: AWS EC2, VPC, AWS Lambda, EBS, S3
• CI/CD: Jenkins, GitHub Actions
• Metrics: Prometheus, Mimir, Grafana, Alertmanager
• Logs: Loki, Vector
• Traces: Tempo, OpenTelemetry, Alloy
• Profiling: Pyroscope
• RUM: Grafana Faro, OpenTelemetry SDK
• Infrastructure as Code: Terraform
• CDN & Edge: Cloudflare, AWS CloudFront
• AWS CloudWatch