Staff Software Engineer, AI Developer Experience

SoFi•San Francisco, CA

About The Position

SoFi’s AI Platform team builds the foundation that enables engineers across the company to deliver software faster and more safely using AI. The AI Developer Experience team owns interactive coding strategy and its infrastructure, including AI economics, coding harnesses and harness vending, an evaluation platform, the enterprise AI Gateway, and intelligent model routing. As a founding Staff Software Engineer, you will provide technical leadership, own major platform systems, establish engineering standards, and collaborate with Autonomous Coding, CI/CD, Code Factory, and other engineering teams. The role involves designing, building, and operating foundational platform components; creating telemetry and cost-allocation mechanisms for AI usage; guiding architecture and execution across teams; and maintaining a 6-to-12-month technical vision. You will work directly with internal developers to understand their needs and improve adoption of interactive coding tools, while supporting operational excellence and mentoring engineers. You will also use AI tools responsibly to accelerate engineering work, validate AI-generated code with automated tests, protect sensitive data and proprietary assets, and help establish effective harnesses, prompt templates, and operational guardrails.

Requirements

Bachelor’s degree plus 3–4 years of experience, or master’s degree plus 2–3 years of experience, in software engineering, computer science, or a related field.
Advanced programming expertise in at least one modern backend language, such as Java, Kotlin, Python, Go, or Scala.
Distributed systems expertise and sound judgment regarding security, concurrency, fault tolerance, horizontal scaling, and idempotence.
Technical leadership experience guiding feature delivery, cross-functional alignment, and architecture on production platforms.
Hands-on experience developing with or building AI-assisted developer tools and LLM-based systems, such as coding assistants, autonomous agents, prompt engineering, or tool execution frameworks.
Hands-on experience with AWS or GCP cloud infrastructure, containerization, modern CI/CD pipelines, and operational on-call service ownership.
Product-oriented engineering mindset, treating internal developers as customers and measuring success through platform adoption and user sentiment.
Clear written and verbal communication skills, including drafting structured technical design documents and presenting options to diverse stakeholders.
Ability to apply AI tools responsibly, validate AI-generated code with automated tests, and uphold security and safety standards.
Ability to comply with data governance and security policies and protect sensitive data and proprietary assets.

Nice To Haves

Direct experience building or operating LLM evaluation pipelines, model proxies, or semantic request routing layers.
Experience engineering internal developer platforms, developer tooling, or CLI workflows.
Experience with cost telemetry, usage-based metering, or FinOps instrumentation.
Background in fintech or adjacent environments with strict data compliance and security regulations.

Responsibilities

Design, build, and operate foundational components of the AI Gateway, model routing, evaluation platform, and harness vending.
Develop usage telemetry, cost allocation, and chargeback mechanisms that provide visibility into AI spending per developer.
Plan feature architecture and implementation paths, parallelize work across teammates, and align cross-team dependencies with product and engineering leaders.
Facilitate design reviews, evaluate architectural trade-offs, and align technical outcomes with the broader engineering community.
Define coding conventions, automated patterns, and architectural standards for maintainable and extensible software.
Automate build, test, and deployment workflows; help resolve production incidents and address root causes through post-incident actions.
Maintain a 6-to-12-month technical vision, anticipate system growth, prioritize architectural hygiene, and make build-versus-adopt decisions.
Work directly with internal developers to understand workflows, drive adoption of interactive coding tools, and incorporate feedback into system enhancements.
Mentor engineers through constructive code and design feedback.
Use AI tools responsibly to prototype solutions, navigate unfamiliar codebases, and accelerate implementation while maintaining security and safety standards.
Review and validate AI-generated code and support implementations with comprehensive automated tests.
Follow data governance and security policies to protect sensitive member data and proprietary assets from unapproved tools.
Share effective harnesses, prompt templates, and operational guardrails to improve team productivity.