Role OverviewWe're looking for a Staff Software Engineer who can define and scale how our engineering organization builds software with AI agents. This role requires deep fluency in 'loop engineering' with tools like Claude Code: designing the feedback loops, guardrails, and workflows that let AI agents safely and effectively contribute to production codebases.
What You Will Do
Design and implement 'loop engineering' workflows using Claude Code and similar tools, architect Model Context Protocol (MCP) servers and tool schemas, define human-in-the-loop checkpoints, autonomy boundaries, and rollback strategies for agent-driven changes, and drive test-driven development practices across teams.
Why It Might Be a Fit
You'll need to understand the full agentic software development lifecycle, from context design through evaluation, deployment, and observability, while also being a hands-on expert across our core stack. This is a staff-level role: you'll set technical direction, define best practices for agentic development across teams, and remain deeply hands-on in architecture and code.
Requirements
- Hands-on experience with Claude Code or comparable agentic coding tools in production settings
- Strong understanding of MCP (Model Context Protocol) building and consuming MCP servers
- Context engineering: managing context windows, prompt caching, and information architecture for long-running agent sessions
- Awareness of prompt injection and agent-specific security risks
- Deep experience with Java microservices
- Strong REST API design skills
- C4 model architecture documentation and system design
- Domain-Driven Design principles
- Node.js and Python for tooling, services, and automation
- MongoDB schema design, performance, and scaling
- ReactJS, headless component architecture, and microfrontend design patterns
- Test-driven development (TDD) as a core practice, not an afterthought
- Playwright for end-to-end testing
- Contract testing (e.g., Pact) across microservice boundaries
- AWS (compute, storage, networking, IAM at minimum — specific services per team stack)
- CI/CD pipeline design, including safe patterns for agent-generated code changes
- New Relic for application observability and performance monitoring
- Solid understanding of how LLMs work capabilities, limitations, failure modes
- Experience with agent orchestration patterns (subagents, delegation, multi-step planning)
- Ability to write clear, actionable RFCs and technical design docs
- Track record of cross-team technical influence without direct reporting authority
- Experience mentoring engineers, including on how to work effectively with AI-assisted development tools
- Comfortable setting technical policy and standards that other teams adopt
Benefits
- Medical, dental, vision, pharmacy, life, accidental death & dismemberment, and disability insurance
- Employee assistance program
- 401(k) retirement plan
- 10 days of paid time off per year
- 10 paid holidays per year
]]>