Staff Software Engineer, AI

About the role

As a Staff Software Engineer, AI, you’ll help define the technical direction of the AI-powered product and platform—designing and building the software systems that enable highly reliable, production-grade AI agents in complex enterprise environments.

This is a hands-on Staff-level engineering role for someone who combines exceptional software engineering fundamentals, deep AI expertise, strong product instincts, and technical leadership. You’ll work across the stack—from distributed backend systems and AI infrastructure to agent orchestration and customer-facing product experiences—while influencing architecture and engineering standards across the organization.

Location: San Francisco, CA — Hybrid

Compensation: Up to $310,000 base + equity

What You’ll Own

Architect & Build AI-Native Software Systems

Design and build scalable software architecture supporting AI-powered products and agentic workflows

Set technical direction for critical components across the AI application and platform stack

Build reliable distributed systems capable of supporting complex, enterprise-grade AI workloads

Identify architectural and engineering bottlenecks impacting reliability, scalability, and development velocity

Make high-leverage technical decisions and remain hands-on through implementation and production

Build & Ship AI Products

Own complex product areas end-to-end—from architecture and implementation through deployment and iteration

Build agent orchestration, retrieval, workflow, and reasoning systems capable of handling real-world enterprise complexity

Develop the infrastructure surrounding LLMs, including model integrations, prompts, tools, guardrails, evaluations, observability, and feedback loops

Build APIs, backend services, data pipelines, and platform capabilities that power AI-native product experiences

Rapidly prototype new approaches, validate them with real users, and harden successful solutions for production

Debug difficult issues spanning application code, AI behavior, infrastructure, and distributed systems

Drive Engineering Standards & Technical Direction

Partner with Engineering, Product, and Design leadership to translate product strategy into scalable technical architecture

Establish engineering patterns and standards for building reliable AI-powered software

Create reusable platforms, abstractions, APIs, and tooling that increase engineering velocity across the organization

Influence technical decisions across multiple teams and product areas

Mentor senior engineers and raise the technical bar through architecture reviews, code reviews, and hands-on collaboration

Balance long-term architectural quality with the speed required to ship and learn

Advance the AI Platform

Stay at the frontier of LLMs, agents, applied AI, and AI-native software development

Rapidly evaluate new models, frameworks, infrastructure, and techniques for practical production use

Improve how AI systems are evaluated, monitored, debugged, and operated in production

Develop internal knowledge and best practices around building reliable AI applications

Help establish the organization as a technical leader in enterprise and vertical AI

Who You Are

You’re a Staff-level software engineer who happens to be deeply experienced in AI—not an AI researcher who occasionally writes production code.

These principles resonate with you:

Software engineer first: You have exceptional engineering fundamentals and know how to build maintainable, scalable production systems

Deep AI fluency: You understand how modern LLM and agent systems work and can translate rapidly evolving AI capabilities into reliable products

Technical leadership: You influence architecture and engineering direction through expertise and credibility rather than authority

Systems thinker: You understand second-order effects and design beyond the immediate feature

Product-minded: You care about whether customers actually receive value, not simply whether the technology works

Multiplier: Your architecture, tooling, mentorship, and technical decisions make other engineers more effective

Principled pragmatism: You know when sophisticated engineering is warranted and when the right answer is simply to ship

End-to-end ownership: You take responsibility for outcomes—from initial architecture through production reliability

Experience

We care more about capability and trajectory than checking every box, but strong candidates will typically bring:

8+ years of production software engineering experience, with significant experience building AI/ML-powered products or platforms

Staff-level experience owning architecture and complex technical systems across multiple teams or product areas

Deep expertise in TypeScript, Python, backend engineering, APIs, and distributed systems

Proven experience designing and shipping LLM-powered applications into production

Experience building agentic systems, orchestration layers, tool-calling workflows, and multi-step AI applications

Strong knowledge of RAG, retrieval architectures, vector databases, embeddings, and data pipelines

Hands-on experience with modern LLM APIs and ecosystems such as OpenAI, Anthropic, Gemini, LangGraph, or similar technologies

Experience designing evaluation frameworks, observability systems, guardrails, and reliability infrastructure for AI applications

Strong understanding of traditional software reliability alongside the unique failure modes introduced by probabilistic AI systems

Experience designing scalable services and systems for enterprise customers

Track record of influencing technical direction and mentoring experienced engineers

What Should Excite You

AI-native product engineering: Building products where AI is fundamental to the architecture rather than an added feature

Enterprise-grade reliability: Turning probabilistic AI capabilities into software professionals can depend on

Agentic systems: Building agents capable of reasoning, retrieving information, using tools, and completing complex workflows

Human-in-the-loop systems: Determining where automation creates leverage and where expert judgment should remain involved

Nuanced evaluation: Measuring quality when there isn’t always a single objectively correct answer

Explainability: Making AI behavior transparent, debuggable, and trustworthy

Complex domains: Building elegant software for environments involving compliance, security, and enterprise rigor

Shipping real value: Moving quickly from prototype to production and building AI experiences customers actively rely on

Benefits

Comprehensive health and wellness benefits

Flexible time off and work schedules

Technology reimbursements

401(k) plan

Twice-yearly in-person offsites across the U.S.

Wellness benefits starting on your first day