Staff Software Engineer, AI Pod

Toast Toast · Enterprise · Dublin, Ireland · R & D : Cloud Service Infra

Staff Software Engineer for Toast's internal AI platform team (AI Pod). The role focuses on building core AI platform infrastructure like LLM proxy, key management, and observability pipelines, as well as architecting and delivering autonomous agents for the SDLC. It also involves developing plugins for an internal marketplace and leading technical design for AI-powered development services. The position requires strong backend service experience, familiarity with AI coding assistants and agentic frameworks, prompt engineering skills, and technical leadership.

What you'd actually do

  1. Design, build, and ship core AI platform infrastructure, including the LLM proxy, AI key management, and observability pipelines powering Toast's internal AI ecosystem
  2. Architect and deliver autonomous agents that participate in the SDLC, including the AI Review GitHub App and Developer Platform MCP integrations
  3. Build on and help maintain the internal plugin marketplace, developing plugins that encode Toast's architectural standards, PR patterns, and quality practices directly into AI assistant behavior
  4. Lead technical design and implementation of MCP (Model Context Protocol) services and no-code service templates that accelerate AI-powered development across Toast
  5. Drive adoption of agentic development practices through tooling, internal evangelism, and hands-on enablement across engineering teams

Skills

Required

  • Designing and implementing scalable backend services
  • Java, Kotlin, or another object-oriented language
  • building scalable backend services
  • building, deploying, or operating LLM-powered agents or AI-assisted developer tooling
  • AI coding assistants (e.g., Claude Code, Cursor, GitHub Copilot)
  • custom plugins, skills, or hooks
  • MCP (Model Context Protocol), tool use patterns, or agentic frameworks
  • prompt engineering skills
  • technical leadership
  • distributed systems
  • API design
  • cloud-native infrastructure
  • customer empathy
  • developer pain points into platform solutions

Nice to have

  • A2A (Agent-to-Agent) protocols
  • multi-agent orchestration patterns
  • machine learning concepts
  • model evaluation
  • ML infrastructure
  • developer experience platforms
  • internal tooling
  • API gateways
  • observability tooling for AI/LLM systems (e.g., Langfuse, DataDog)

What the JD emphasized

  • autonomous agents that participate directly in the software development lifecycle
  • AI infrastructure
  • LLM Proxy
  • autonomous agents
  • AI Review GitHub App
  • Developer Platform MCP integrations
  • agentic development practices
  • tool use patterns
  • agentic frameworks
  • prompt engineering skills
  • LLM behavior changes with context, instructions, and tool definitions

Other signals

  • building AI infrastructure
  • autonomous agents
  • LLM Proxy
  • developer tooling