Staff Software Engineer, AI

Role overview Lead the technical direction of an AI platform team building agents that plan and carry out complex digital media operations. The role spans agent architecture, orchestration, integrations, evaluation, and reliability, while helping other engineering teams adopt effective AI development practices.

Responsibilities - Set technical direction for the agent platform and guide its architecture and implementation. - Build orchestration, tools, and evaluation systems that help language model agents operate reliably in a complex domain. - Expose platform capabilities to agents through protocols such as MCP and A2A. - Establish reusable patterns and practices for agentic and AI-assisted engineering across teams. - Mentor engineers and explain technical decisions to engineering and product partners. - Use AI coding tools in day-to-day development, taking responsibility for the quality of generated work.

Requirements - At least 8 years of professional software engineering experience, including recent production work on LLM applications or agent systems. - Strong technical leadership, including ownership of a team or major system’s direction and influence beyond its immediate scope. - Expert Python skills and experience designing production APIs and backend services, including protocol integrations. - Deep experience building and operating production agent systems, with asynchronous or event-driven patterns, prompt design, structured outputs, and reliability techniques. - Experience creating evaluation strategies and using observability tools for AI systems. - Ability to mentor engineers and communicate complex systems clearly; sound judgment about balancing speed and quality.

Nice to have - Experience with AWS Bedrock AgentCore, including Runtime, Gateway, or Identity. - Designing MCP tool sets or skills libraries used across teams or by external partners. - Agent authentication and authorization, including OAuth flows or secure token handling. - Experience with retrieval-augmented generation, vector databases, or related knowledge retrieval methods. - Familiarity with Python frameworks and tools such as LangChain, LangGraph, FastMCP, FastAPI, or Langfuse.

Benefits and work setup Flexible work options include remote, hybrid, or office-based work in the U.S. or Canada. Listed benefits include flexible scheduling, retirement plan matching, mental health support, paid sabbaticals, and parental leave.