Systems Development Engineer

ABOUT US

At Union, we are solving one of the hardest challenges in AI infrastructure today: enabling high-velocity iteration while maintaining seamless production-readiness for AI workloads at scale.

Flyte https://flyte.org/, the open-source project we steward, has emerged as the modern standard for data and AI orchestration, and is trusted by leading technology organizations including LinkedIn, Stripe, and Wayve to run millions of mission-critical workflows on the platform. These workflows comprise data preparation, model training, and scaled inference spanning thousands of GPUs, all major clouds, and on-premise infrastructure.

We have a technical founding team who created Flyte while at Lyft, a deep bench of infrastructure experts from top companies, and have raised from top investors like NEA and Nava Ventures.

ABOUT THE ROLE

We are hiring a Systems Development Engineer to improve the reliability, operability, and customer experience of our production platform. This is a 50/50 operations and engineering role. Part of your time will be spent investigating customer-impacting production issues, and part will be spent building the tools, automation, system design, and engineering practices that prevent those issues from recurring.

For this role, production is the customer. You will work from real operational signals: customer issues, incidents, on-call pages, recurring support patterns, and gaps in observability or automation. You will sit in engineering and partner with customer-facing teams to turn those signals into durable platform improvements.

This is not a traditional support role. It is a systems engineering role for someone who can debug deeply, communicate clearly, and write software that reduces operational load. The full engineering team backs you on on-call.

This role is hybrid, based out of our Seattle office.

WHAT YOU'LL DO

• Investigate and resolve customer-impacting production issues across cloud infrastructure, workflow execution, access control, storage, networking, deployment systems, and observability.

• Identify patterns in customer issues and convert them into automation, product improvements, runbooks, tests, or design changes.