Principal AI Platform Engineer

Hi Marley • Boston, MA

1d • Hybrid

About The Position

Hi Marley is seeking a Principal AI Platform Engineer to be a foundational hire on their new AI Operations team. This role involves building the core infrastructure for their internal agentic workforce, including the AWS hosting layer, identity model, accountability registry, agent registry, and tool catalog. The engineer will partner with the CTO and a UX-focused builder to make Hi Marley an AI-native company, operating on a strict AWS stack for auditability. The position requires deep AWS expertise, fluency in identity and authorization, strong product instincts for developer experience, AI-forward thinking, and excellent communication skills. This role is based in Boston, requiring 2-3 days per week in the office.

Requirements

• Deep AWS expertise: ability to architect a multi-tenant platform on AWS, including Bedrock, Cognito, IAM, EventBridge, Lambda, DynamoDB or Postgres, and IaC (CDK or Terraform).
• Identity & authorization fluency: experience with systems where authorization is architectural, including OAuth 2.1, scoped tokens, role inheritance, and least-privilege design.
• Strong product instincts: ability to think about the developer experience and design user-friendly SDKs or registry schemas.
• AI-forward: experience shipping real LLM-backed systems in production, with opinions on token cost, latency, evaluation, observability, and determinism.
• MCP / agentic systems literacy: familiarity with Model Context Protocol or comparable agent-tool patterns, understanding of its purpose and security boundaries.
• Excellent communicator: ability to present platform work to various audiences and write effective documentation.
• Architectural integrity: preference for building the right primitive on AWS over reaching for higher-level abstractions.
• Fluent in Python, TypeScript, or Go.

Nice To Haves

• Bonus AWS expertise: AgentCore, SageMaker, Workshop Studio.

Responsibilities

• Build the AI Operations platform on AWS, owning the hosting, deployment, and infra-as-code layer for internal agents and tools.
• Architect and build the Accountability Registry (people, ownership, roles), Agent Registry (identity, permissions, data scope, audit trail), and Tool & App Catalog (discover, register, approve).
• Solve the identity-inheritance problem by designing and shipping patterns using Cognito, IAM, and scoped tokens for auditable agent actions.
• Build the developer experience for internal agent builders, including templates, SDK helpers, deploy paths, observability, and cost tracking.
• Partner with the Compliance & IT lead on security tooling, implementing scans, audits, MCP governance hooks, and DLP signals.
• Evangelize the platform internally through lunch-and-learns, internal docs, demo droplets, and office hours.
• Utilize AI in daily workflows, defaulting to the question 'could an agent do this?' before writing code.

Benefits

• Equity grants for all employees
• A 4% matching 401(k) program
• Medical, dental, vision, disability, and life insurance coverage for employees working 30+ hours per week
• Monthly wellness stipend
• Paid parental leave
• A flexible vacation policy

What This Job Offers

• Job Type: Full-time
• Career Level: Principal
• Education Level: No Education Listed