Senior AI Backend Engineer
San Francisco (HQ) • FullTime
Posted 5mo ago
About the job
We are seeking experienced Backend and Applied AI Engineers to develop the core systems powering Within's multimodal agents for enterprise environments. This role is at the intersection of distributed systems and applied AI, focusing on designing and shipping backend services that handle real-world data ingestion, orchestrate AI model and agent workflows, ensure enterprise-grade security, and deliver reliable outcomes at scale. You will collaborate closely with product and design teams to transform ambitious AI capabilities into production-grade features for Fortune 500 enterprises. A strong emphasis is placed on engineering fundamentals such as correctness, observability, security, and performance, balanced with rapid iteration speed. You will also play a key role in setting technical direction for critical subsystems and mentoring junior engineers.
Responsibilities
- Design, experiment, build, test, and deploy backend services for the AI platform, owning critical infrastructure end-to-end.
- Implement AI orchestration pipelines that ensure reliable agent behavior at scale, incorporating structured outputs, tool calling, state management, and grounding.
- Build comprehensive evaluations and regression tests for AI model changes.
- Define and standardize orchestration patterns and evaluation frameworks.
- Make foundational decisions on technology choices, data models, and service boundaries.
- Balance long-term architectural vision with a rapid shipping cadence.
Requirements
- 4+ years of experience designing and shipping production-grade products.
- Proficiency in designing and building reliable, observable, secure, and fast backend services, APIs, data models, and infrastructure.
- Strong systems thinking and foresight to anticipate future service evolution.
- Experience designing at least one system from scratch that is used in production at scale.
- Experience shipping multiple AI-powered features to production, with opinions on orchestration patterns, eval frameworks, and failure modes.
- Familiarity with AI-native development across the SDLC, including tools like Cursor, Claude Code, and Copilot.
- Ability to balance shipping speed with quality and craft.
- Ability to create leverage by unblocking others through code reviews, architecture guidance, and technical writing.