Software Engineer, CI Platform Infrastructure

Airbnb

2d•$162,000 - $190,000•Remote

About The Position

The Developer Infrastructure organization’s mission is to enable Airbnb engineers and AI Coding Agents to ship high-quality software with confidence and speed. We work on infrastructure that accelerates Airbnb’s engineering productivity across all stages of the software development lifecycle. This role will work on our CI platform infrastructure on the CI+Build team. This is potentially one of the highest-leverage investments for our engineering organization as we accelerate the iteration loop for engineers and AI coding agents that need a scalable workload orchestration platform to create, build, test, review, and merge PRs. We are reimagining our CI platform's workflow orchestration to handle AI-scale software development. You will be doing feature-development work on a green-field opportunity to modernize scheduling, queueing, backpressure, consistency, caching, fault tolerance, and observability across our CI platform infrastructure components. Working alongside experienced Staff and Senior Staff engineers, you will help design and build key pieces of this new platform, with real ownership over the components you work on. There is significant room for impact in how we schedule work onto the most cost-efficient compute stack, and you will have a direct role in shaping that. Note that our team does not focus on writing/maintaining the application logic tests that run on our CI platform.

Requirements

Bachelor's and/or Master's degree, preferably in CS, or equivalent experience.
Interest in working on distributed systems engineering and optimization problems.
General knowledge of most of the following: k8s, EC2, Golang, Docker.
Metrics-driven approach to decision-making, with an interest in delivering quantifiable impact.

Responsibilities

Experimentation and prototyping new features in our remote sandbox development environment to enable new capabilities in CI-related workflows.
Debugging workflow orchestration queueing delays and measuring the performance of our remote execution/cache infrastructure to improve correctness, reproducibility, and speed.
Identifying root causes of infrastructure-related reasons for test flakiness in CI.
Optimizing the speed of autoscaling our CI workers to handle a burst in CI workloads from our product engineering teams.
Up-leveling the observability and alerting of our CI platform across build systems, flaky test management systems, and our merge queue.
Building an improved canarying mechanism that will let us roll out workflow orchestration changes iteratively and safely.