Platform Engineer (Agent Runtime)

📋 Description

• Own the agent execution environment health: isolate sessions and resources
• Debug prod vs test: work with Agent/QA to root cause issues
• Tune cold-start and concurrency for critical-path functions
• Understand session isolation models and platform guarantees
• Build observability: tracing, drift detection, signals
• Track per-agent/session cost and flag expensive operations
• Deploy and manage runtime infra via CI/CD; IAM roles and configs
• Test and improve isolation tests to prevent cross-agent impact

🎯 Requirements

• 5+ years in platform, SRE or infra role with production AWS workloads
• Hands-on with AWS observability tools (CloudWatch/X-Ray) for root cause
• End-to-end IaC deployment experience (Terraform/CDK) in CI/CD
• Understanding of compute isolation concepts containers/microVMs
• Cost/performance debugging; trace slow workloads to root cause
• Strong incident response: calm, methodical root-causing
• Direct experience with AWS Bedrock or SageMaker
• Cold-start/concurrency tuning experience (Lambda/dedicated pools)
• Experience in regulated environments with formal change reviews
• Familiar with AI token pricing and model tiering