Staff Software Engineer, Backend (Capacity Modeling)

Affirm is reinventing credit to make it more honest and friendly, giving consumers the flexibility to buy now and pay later without any hidden fees or compounding interest.

The Capacity Modeling team ensures Affirm can safely handle forecasted traffic—especially during peak sales events—by translating demand forecasts into concrete, testable capacity plans. They build and maintain statistical capacity models that map traffic inputs (e.g., checkout volume, app opens, batch/job activity) to resource outputs (CPU, memory, connections) and then convert those forecasts into recommended capacity settings such as pod counts, HPA ranges, and/or infrastructure sizing changes across regions and environments.

The coverage of infrastructure resources includes not only Compute but also infrastructure such as ElasticCache, DynamoDB, AuroraDB, and more.

What You'll Do

You will be responsible for setting technical strategy for your team on a year-long time scale, and help your team tie it together with critical, business-impacting projects.

You will collaborate across teams in the product development lifecycle by collaborating with product management, design & analytics to ensure technical sustainability, risks and trade-offs are well understood and managed.

You will act as a force-multiplier for your team through your definition and advocacy of technical solutions and operational processes.

You take ownership of your team’s operations and availability by ensuring you have the right monitoring, triage rotations, playbooks, policies, testing and alerting in place to support “keep the lights on” & on-call efforts.

You will foster a culture of quality and ownership on your team by setting code review and design standards for your team, and advocating for them beyond your team through your writing and tech talks.

You will help develop talent on your team by providing feedback and guidance, and leading by example.

What We Look For

You have 8+ years of experience designing, de