Senior AWS AgentCore Platform Engineer

About

Experience Desired: 10+ Years.

Qualification:

We are seeking a Senior AWS AgentCore Platform Engineer to design and scale observability, cost management, monitoring, and security frameworks for next-generation agentic AI workflows on AWS. This role will focus on distributed tracing, LLM observability, cost optimization, and governance across AgentCore platforms, MCP servers, and AWS Bedrock integrations.

Responsibilities

• Observability & Distributed Tracing
  • Perform gap analysis across AWS CloudWatch, X-Ray, Bedrock logging, and AgentCore traces
  • Design and implement end-to-end observability using Dynatrace
  • Build post-deployment validation pipelines for agents and MCP servers
  • Implement distributed tracing & structured logging for:
    • LLM decision flows
    • Tool usage
    • Sub-agent interactions
  • Evaluate LangFuse, LiteLLM proxies, and AWS-native tools and define target architecture
• Cost Tracking & TCO
  • Enhance AWS tagging taxonomy for:
    • Agent runtimes
    • MCP servers
    • Vector databases
    • Bedrock token usage
  • Build granular cost models by team, department, and workload
  • Develop CloudWatch dashboards & AWS Budgets alerts
  • Automate cost reporting (Email + Microsoft Teams) with anomaly detection
• Monitoring & Incident Management
  • Deployment failures
  • Runtime errors
  • Tool invocation failures
  • MCP connectivity issues
  • Integrate alerts with Microsoft Teams & Email
  • Create and maintain runbooks in Confluence
  • Evaluate AWS-native vs third-party monitoring tools
• Security & Governance
  • Conduct IAM and tagging strategy risk assessments
  • Evaluate Cedar policy engine (AgentCore) for fine-grained access control
  • Design ABAC (Attribute-Based Access Control) architecture
  • Build scalable Terraform modules for identity and access management

Key Skills

• Cloudwatch
• X-Ray
• Dynatrace
• LLM platforms
• Terraform