Forward Deployed Engineer (SRE Agent) - Staff Engineer

PagerDuty, Inc. (NYSE: PD), the global leader in AI-first digital operations, is seeking a Forward Deployed Engineer (SRE Agent) to join their customer-focused SRE Agent team. As a Forward Deployed Engineer, you will work inside PagerDuty’s largest customer accounts, connecting the SRE Agent to their production environments (observability tools, infrastructure as code, GitOps tooling, and CI/CD pipelines) and guiding their teams from AI-assisted incident response to supervised and autonomous actions with safety guardrails.

You will also contribute code directly back to the core SRE Agent product, building reusable tools, connectors, and skills derived from real customer needs.

This is an exciting opportunity to join the product engineering team as the Staff Engineer and directly shape how agentic AI operates within the most strategic customers. The ideal candidate brings a strong background as a software engineer, hands-on experience shipping LLM/agentic systems to production, and the customer-facing skills to bridge technical execution with business strategy.

Key Responsibilities

• Embed and drive SRE Agent adoption: Work onsite and in regular contact with 2–4 strategic customer teams to deploy the SRE Agent, configure tools and connectors, and embed AI actions across their incident management lifecycle to drive adoption and value.
• Build and contribute core agent capabilities: Scope, design, and build reusable agent tools, connectors, and skills on the agent’s extensibility surfaces, shipping them through the product team’s repos, review process, and evaluation gates.
• Run pilots and unblock customer adoption: Stand up new agentic capabilities with strategic accounts first, gather evaluation data, identify technical blockers (permissions, telemetry, missing connectors), and build production fixes directly into the product.
• Integrate with production stacks and progressive delivery: Connect the SRE Agent to IaC (Terraform, Pulumi), GitOps (ArgoCD), CI/CD pipelines, and observability platforms, guiding customers safely from AI-assisted to autonomous operations.
• Set technical direction and feedback insights: Define 3–6 month technical roadmaps across assigned accounts, provide architectural guidance to product managers and engineering managers, and bring customer friction and evaluation data back to the core team.

Basic Qualifications

• 8+ years of experience building and shipping production software across team boundaries (assessed against PagerDuty Staff Engineer / SDE V expectations).
• Hands-on experience with LLM and agent systems in production, including agent development, tool-use frameworks (MCP or equivalent), evaluation pipelines, and guardrail design.
• Customer-facing skills. You can run a technical design session with a customer’s engineers and present the business case to their leadership.
• Comfortable with ambiguity. You can start from a vague problem statement in an unfamiliar environment and get to working software without much direction. Former founders and founding engineers tend to do well in this role.
• A record of shipping software that got adopted, and the judgment to recognize which customer requests generalize and when the right answer is not to build.
• Infrastructure & Deployment expertise: Demonstrated experience with Terraform or Pulumi at scale, GitOps (ArgoCD or equivalent), CI/CD pipeline design, and progressive delivery patterns (canary, blue/green, soak periods).
• Willingness to travel 25–50% during active customer engagements.

Preferred Qualifications

• Practical operational knowledge of incident management and observability concepts (on-call, postmortems, SLOs/SLIs, alerting).
• Customer-facing technical consulting, forward-deployed engineering, or solutions architecture experience on top of a core software engineering background.
• Background as a founder or founding engineer accustomed to navigating high ambiguity and driving unscripted problem-solving.
• Experience with policy-as-code, fine-grained permissioning, or safety guardrail patterns for agent-driven automation.
• Production experience with PagerDuty or competing digital operations and incident management platforms.
• Active open-source contributions to the AI agent or Model Context Protocol (MCP) ecosystem.