Senior AI Reliability Engineer | Azure & SLOs

EPAM Systems; Warszawa; Hybrid; PLN 240,000 - 360,000; Full time

Benefits offered by this job

Hybrid work model

Relocation opportunities

Work abroad opportunities

Health insurance

Employee stock purchase plan

Job summary

EPAM Systems seeks a Senior Site Reliability Engineer to own reliability and observability for production AI systems in a hybrid model. You will own deployment end-to-end on Azure, implement infrastructure as code, and ensure environments are rebuildable from source.

You will build LLM-aware observability with traces on model calls, cost and latency dashboards, and define SLOs. You will manage on-call rotations, incident responses, and security postures across the lifecycle.

Qualifications

• 5+ years operating cloud production systems.
• Experience in Azure IaaS/PaaS operations.
• Proficiency in observability stacks with LLM tracing and Python/Bash automation.
• Knowledge of FinOps basics for AI workloads.
• Familiarity with incident triage, runbook drafting and automation.

Responsibilities

• Own deployment end-to-end on Azure with IaC, CI/CD, and environment management.
• Build LLM-aware observability with traces, eval sampling, drift detection, and dashboards.
• Define and defend SLOs for availability, latency and quality.
• Run incident management with on-call models and blameless postmortems.
• Monitor token economics, budgets, and capacity for AI workloads; ensure security posture.

Skills

Cloud operations

SLOs and on-call

Python scripting

FinOps for AI

Incident management

Observability

Automation

Cost optimization

Tools

Azure

Terraform

Bicep

CI/CD tooling

Kubernetes

Job description

EPAM Systems seeks a Senior Site Reliability Engineer to own reliability and observability for production AI systems in a hybrid model. You will own deployment end-to-end on Azure, implement infrastructure as code, and ensure environments are rebuildable from source.

You will build LLM-aware observability with traces on model calls, cost and latency dashboards, and define SLOs. You will manage on-call rotations, incident responses, and security postures across the lifecycle.