Principal AI Platform Engineer

Responsibilities

• Own hosting, deployment, and infrastructure-as-code for Hi Marley’s internal agents and tools on AWS.
• Design and build the Accountability Registry, Agent Registry, and Tool & App Catalog for agentic governance.
• Implement deterministic, auditable identity inheritance using Cognito, IAM, scoped tokens, and least-privilege authorization.
• Build templates, SDK helpers, deployment paths, observability, and cost tracking for internal agent developers.
• Partner with Compliance and IT on security scans, audits, MCP governance hooks, and DLP signals.
• Evangelize the platform through documentation, demonstrations, lunch-and-learns, and office hours.

Requirements

• Deep AWS expertise, including Bedrock, Cognito, IAM, EventBridge, Lambda, DynamoDB or Postgres, and CDK or Terraform.
• Experience architecting multi-tenant AWS platforms and designing identity, authorization, scoped-token, and role-inheritance systems.
• Production experience shipping LLM-backed systems, with understanding of token cost, latency, evaluation, observability, and determinism.
• Familiarity with Model Context Protocol or comparable agent-tool patterns and their security boundaries.
• Ability to design developer-facing SDKs and registry schemas with strong product and developer-experience instincts.
• Fluency in Python, TypeScript, or Go.
• Excellent communication skills for presenting to engineers, executives, and customers and writing effective documentation.
• Curiosity about AI and emerging technologies, with sound judgment in applying them responsibly.
• AgentCore, SageMaker, and Workshop Studio experience are preferred.

Benefits

• Equity grants for all employees.
• 4% matching 401(k) program.
• Medical, dental, vision, disability, and life insurance coverage for employees working 30 or more hours per week.
• Monthly wellness stipend.
• Paid parental leave.
• Flexible vacation policy.
• Flexible hybrid work environment with three days per week in the Boston office.