AI Agent Performance Lead

Rivian•Irvine, CA

About The Position

Rivian is refining its operating model for enterprise AI agents that engage customers intelligently at every milestone — from first inquiry through purchase and ownership. This role serves as the platform owner and architect for conversational AI, defining, testing, evaluating, and optimizing agent behaviors to drive measurable sales outcomes. Across the customer org, setting org-wide strategy and standards across multiple teams/platforms with broad authority and being recognized as the functional authority beyond one domain. Embedded within the Sales organization, this role bridges sales enablement and AI execution. You will own the conversation logic, customer-journey mapping, and performance optimization of AI agents across customer touchpoints. This role is distinct from backend engineering. Commercial Technology owns platform architecture, API integrations, security, and data access. You will translate those capabilities into sales value by designing agent behaviors, improving prompts and tool-calling logic, evaluating agent performance, and iterating based on commercial outcomes. At the intersection of sales strategy and engineering, you will analyze conversation transcripts to identify failure points, investigate LLM reasoning and agent behavior, recommend improvements to agent instructions, and architect the measurement framework that demonstrates impact. Ultimately, you will establish the discipline, measurement, and cross-functional operating model required to scale Sierra AI as a trusted, high-performing sales platform.

Requirements

7+ years of technical sales/revenue operations experience, or software engineering, or AI operations, with proficiency in TypeScript or JavaScript (Node.js).
Hands-on experience evaluating LLMs, understanding how models reason and fail.
Comfortable reading REST API specs and working with JSON Schemas provided by technical teams.
Ability to translate business goals (lead qualification, response time reduction, conversion) into deterministic agent code.
Deep understanding of sales funnels, customer journeys, frontline workflows, and the operational levers that influence conversion and customer satisfaction.
Strong executive presence required with experience successfully presenting and distilling complex concepts for a C-suite audience.
Experience explaining technical concepts (prompting, tool calling, model behavior) to non-technical stakeholders and vice versa.
Exceptional conversational writing skills: able to craft concise, natural, and persuasive dialogue that aligns with brand tone and voice.
Demonstrated ability to define product strategy and translate ambiguous business problems into prioritized roadmaps and executable requirements.

Nice To Haves

Experience with agent frameworks (LangChain, CrewAI, or similar) or building multi-step reasoning systems.
Background in product operations, sales engineering, or customer success—roles where you've had to understand both the technical side and the commercial impact.
Familiarity with sales tech stack tools (e.g., Salesforce, HubSpot, Gong, Outreach, Chili Piper).
Experience with state machine design or conversational UX frameworks.
Basic working knowledge of Python for data processing or evaluation scripting.
Exposure to automotive, retail, or direct-to-consumer sales environments.

Responsibilities

Define conversation logic, decision policies, and behavioral guardrails (including safety, privacy, and brand tone) for AI agents across customer scenarios and journeys, enabling reliable, effective behavior at scale.
Map customer journeys through the agent: define decision points, routing logic, qualification criteria, and handoff conditions that optimize for lead quality, conversion, and customer experience.
Define the metrics that matter for agent performance: conversation completion rates, lead quality scores, customer satisfaction, routing accuracy, time-to-resolution, and impact on sales pipeline.
Partner with Analytics/Commercial Technology to build dashboards that surface agent performance at multiple levels: daily operations view, weekly performance trends, and monthly business impact analysis. Ensure data pipelines capture the signals needed to measure agent effectiveness.
Present channel performance, risk, and roadmap updates in a recurring cadence (e.g. weekly ops review, monthly business review), translating technical detail (model behavior, evaluation results, experimentation outcomes) into clear, decision-ready narratives for an executive audience.
Diagnose model behavior, tool-use failures, instruction conflicts, and workflow breakdowns using transcripts and evaluation data.
Run A/B tests on agent behavior changes; measure impact on conversation quality, customer satisfaction, lead routing accuracy, and sales outcomes.
Treat agent optimization as a continuous process, not a one-time build. Conduct regular analysis of agent conversations: sample and review transcripts to identify patterns, assess quality, understand failure modes, and surface improvement opportunities. Translate agent limitations into product requirements; prioritize based on business impact.
Serve as the primary escalation point for agent-related issues, coordinating cross-functional triage and root cause analysis.
Communicate with field teams (sales, customer support) and cross-functional stakeholders to understand what improvements would have the most impact.
Drive progress in ambiguous, cross-functional environments and remove barriers to execution through clear decisions, ownership, and communication.