Staff Software AI Engineer

Rivian’s Applied AI team is building internal AI applications powered by large language models and agentic workflows. In this Staff Software AI Engineer role (onsite in Palo Alto, CA), you will define the technical direction for an agent platform that integrates enterprise knowledge and tools, with a focus on reliability, governance, and scalable production delivery.

What you’ll work on

You will establish and own the agent platform architecture used to orchestrate long-running, multi-step agents. This includes persistent state management, tool integration, and permission-aware access to enterprise data. You will also build evaluation infrastructure that gates production releases, helping ensure internal AI applications meet quality and performance expectations over time.

• Define and own agent platform architecture for orchestration, persistent state, tool integration, and evaluation-driven production gating
• Create reusable patterns, interfaces, and engineering standards so capabilities can be adopted across the organization
• Set the bar for reliability, observability, and evaluation of AI systems, ensuring production apps meet targets for quality, latency, and cost
• Lead design and implementation of the most complex components, deliver reference implementations, and resolve critical integration and reliability challenges
• Evaluate emerging models, frameworks, and tools, leading build-versus-buy and vendor selection decisions and translating insights into a roadmap
• Partner with Security, Legal, and Compliance to embed authorization, auditability, and data governance into the platform
• Raise the engineering bar through design and code reviews, mentoring, and participation in hiring

Requirements

• 8+ years of professional software engineering experience, including technical leadership of complex production systems
• Experience shipping and operating LLM-based or agentic applications with real users in production, supported by measured quality, adoption, and business impact
• Deep knowledge of agent system design: orchestration, tool calling, retrieval and grounding, context and memory management, durable execution, verification, and evaluation of nondeterministic systems
• Backend and distributed systems expertise: APIs, asynchronous execution, state management, fault tolerance, and observability at scale
• Expert-level Python and experience deploying cloud services with containers and automated delivery pipelines
• A record of setting technical direction and engineering practices adopted beyond your team, with demonstrable impact on systems and engineers
• Experience integrating with enterprise systems and data under security, privacy, and permission constraints
• Ability to translate ambiguous needs into practical technical solutions, with clear communication to engineers, domain experts, and executives
• Bachelor’s, Master’s, or Ph.D. in Computer Science, Engineering, or a related field, or equivalent practical experience

Preferred qualifications

• Experience with multi-agent systems including verification or adversarial checking, persistent state, and recovery or human-approval mechanisms
• Experience with multimodal applications, including document or engineering-drawing understanding, vision models, or computer-use agents
• Familiarity with product development tooling: CAD, PLM, simulation, CAE, or physics-informed machine learning
• Experience improving AI systems through failure analysis, feedback loops, model adaptation, or fine-tuning
• Experience building AI infrastructure or capabilities adopted by multiple teams, or forward-deployed experience at an AI lab or agent-infrastructure company

Technologies: Python, containers