3 days ago
Remote, Poland or Kraków, PolandMid Level
Responsibilities
- Build and maintain observability for AI workloads, including telemetry, dashboards, alerts, SLO/SLI tracking, and improvements when reliability targets are missed.
- Write automation and tooling to reduce operational toil, improve deployment safety, and accelerate incident response.
- Integrate AI workloads into incident-management processes, create runbooks, participate in on-call rotations, and conduct blameless post-mortems.
- Build and maintain CI/CD integrations, deployment safety checks, and rollback automation.
- Collaborate with product engineering teams on reliability improvements, architecture decisions, and operational readiness for releases.
- Contribute to capacity planning, autoscaling configuration, and workload scheduling for AI compute infrastructure.
Requirements
- Expertise in SRE, infrastructure, or platform engineering with extensive operational experience managing large-scale distributed systems.
- Expertise in Kubernetes and large-scale containerization systems.
- Experience defining SLOs and using observability tools such as Prometheus, Grafana, and distributed tracing.
- Proficiency in Python or Go for automation, CI/CD pipelines, deployment safety, and infrastructure-as-code such as Terraform.
- Interest in or experience with AI/ML infrastructure, model serving, or GPU workloads.
- Ability to resolve issues independently, maintain accountability, develop automation and monitoring, and collaborate with engineering teams unfamiliar with SRE practices.
Benefits
- Career development opportunities through programs such as GROW and Mentoring.
- Internal events including the APEX Expo.
- Learning resources including LinkedIn Learning.
- Opportunities to build new skills, explore new roles, and pursue different career opportunities within Akamai.
Tech Stack
Categories
Site Reliability
About Akamai
Akamai builds content delivery, cloud security, and edge compute services for enterprises that run consumer and business web, API, and video workloads. The public company (NASDAQ: AKAM), founded in 1998 and headquartered in Cambridge, MA, operates a global CDN and security platform and sells via subscriptions and usage-based services. It expanded into cloud infrastructure by acquiring Linode in 2022, offering IaaS and managed Kubernetes.
