What You’ll Do
As a Senior Software Engineer on the Infrastructure AI team, you will design, build, and operate goal-oriented AI agents that automate SRE, DevOps, and DBA workflows across Airwallex’s global infrastructure. You won’t be doing operations work yourself — you’ll be building the AI systems that do it. Your agents will autonomously investigate incidents, provision infrastructure, manage databases, execute deployments, and handle the operational tasks that traditionally require human intervention.
You will work on top of Quartermaster, Airwallex’s internal agent platform, building domain-specific agents that interact with real infrastructure tooling (Kubernetes, Terraform, cloud APIs, databases, monitoring systems). This role requires you to think deeply about agent architecture — planning and reasoning loops, tool selection and execution, safety guardrails, human-in-the-loop escalation, and how to make autonomous systems reliable in production infrastructure environments.
This hybrid role is based in Seattle, WA.
Responsibilities
- Build goal-oriented infrastructure AI agents: Design and implement autonomous agents on the Quartermaster platform that handle SRE, DevOps, and DBA workflows — including incident investigation and remediation, infrastructure provisioning, deployment orchestration, database operations, and capacity management.
- Design agent architectures for safety and reliability: Build robust agent loops with planning, reasoning, tool-use, error recovery, and human-in-the-loop escalation. Infrastructure agents touch production systems — they must be safe, auditable, and predictable.
- Integrate agents with infrastructure tooling: Give agents the ability to interact with real systems — Kubernetes, Terraform, cloud APIs (AWS, GCP, Aliyun), monitoring/observability platforms, databases, and CI/CD pipelines — through well-designed tool interfaces.
- Evaluate and improve agent performance: Define metrics for agent effectiveness (task completion rate, accuracy, time-to-resolution, escalation rate). Build evaluation frameworks to measure whether agents reliably achieve their goals.
- Collaborate cross-functionally: Partner with SRE, DevOps, and platform engineering teams to identify the highest-value workflows for agent automation. Understand real operational pain points and translate them into agent capabilities.
- Shape the future of infrastructure AI at Airwallex: Contribute to the technical vision for how AI agents will transform infrastructure operations. Stay at the forefront of agentic AI developments and bring new ideas into the team.