About this role
At Affirm, we exist for the moments that matter—giving people a clear, predictable way to pay over time, with no hidden fees, no surprises, and no tradeoffs on what matters most.
Affirm's engineering team is building a large-scale, fault-tolerant, and global infrastructure that is shared across multiple financial products, merchants, and vendors. Ensuring that our infrastructure is transparently available to all engineers is critical to the success of the business. We pride ourselves on our culture across engineering of writing blameless postmortems as well as writing detailed tech specs and gathering feedback before large changes to systems.
The AI Platform team owns Affirm's paved path for building, running, evaluating, and governing AI systems. Our customers are not only Affirm users, but also every engineer and a growing set of production workloads at Affirm. As a team, we are responsible for building and exposing simple, governed interfaces to models, tools, and agents, with identity, cost attribution, reliability, and security built in. Examples of components we own include the LLM Gateway that fronts every approved model (Claude, ChatGPT, and open-weight models across AWS Bedrock and other providers), the MCP Gateway that mediates and audits tool calls, the Agent Runtime that gives agentic workloads bounded execution, and the evals and observability layer that tells teams whether an AI change is better, cheaper, or safe to ship. We are early: the gateway is in daily use across engineering, the first production agent is live, and the next year takes us from developer productivity to production inference and agents.
What You'll Do
• Plan, execute, and own technical decisions and projects within the AI Platform team, from design through launch and on-call.
• Design and build the gateway, runtime, and evaluation services that take AI workloads at Affirm from pilot to production, securely scaling across millions of users and their transactions.
• Make AI spend attributable and governable: build the identity, quota, routing, caching, and failover controls that let teams adopt new models and tools without re-solving the same platform problems.
• Partner with security, infrastructure, and product engineering teams to set the guardrails for agentic workloads and the on-behalf-of identity model for tool access.
• Create and monitor metrics for your team's systems, and own the operational health of what you ship: monitoring, alerting, runbooks, and on-call.
• Set and improve code review and design standards, and advocate for them beyond your team through writing and tech talks.
• Use AI tools and agentic workflows every day as a multiplier for your own code, design review, and operations, and push the team on what a small AI-native team can deliver.
• Educate, mentor, and grow engineers in the team, and drive best practices around design, coding, and testing.
What We Look For
• You have 4+ years of experience designing, developin