Software Engineering Manager, Business AI Agent

Google•Mountain View, CA

8d•$207,000 - $300,000

About The Position

The Software Engineering Manager leads a team building the AI platform behind Google’s Agentic Commerce strategy, enabling merchant-consumer interactions across Search, YouTube, Ads, and Messaging. The role owns technical direction for applied AI, low-latency scoring, and automated evaluation, while managing engineers, project goals, and product strategy. The team develops reliable, responsive agent experiences, merchant knowledge systems, and tools that support measurable commerce outcomes for businesses worldwide. The manager leads engineering for the Merchant Knowledge Library, retrieval-augmented generation (RAG), autonomous execution tooling, and integrations with the Universal Commerce Protocol (UCP). The role also scales LLM evaluation and AutoRater frameworks to assess conversation quality and factuality, and partners with Product, Data Science, UX, and Google platform teams on datasets, dashboards, and model upgrade standards. The position requires a bachelor’s degree or equivalent practical experience, substantial backend-focused software development experience, expertise in LLM evaluation or reliability measurement, and experience managing and developing software engineers. The listed U.S. compensation range is $207,000–$300,000, plus a 20% bonus target, equity, and benefits.

Requirements

Bachelor’s degree in Computer Science, a related technical field, or equivalent practical experience.
8 years of experience coding in one or more general-purpose programming languages, such as Python, Java, C++, or Go, focused on backend system architecture.
Experience building automated LLM evaluation pipelines, automated rating systems (AutoRaters), or defining LLM-specific reliability metrics, such as grounding, helpfulness, safety, and hallucination or error reduction.
Experience with backend, distributed systems, or AI/ML systems infrastructure.
Experience managing software engineers, including performance management, mentoring, and team growth.

Nice To Haves

Master’s degree or PhD in Computer Science or a related technical field.
3 years of experience working in a complex, matrixed organization.
Experience developing, calibrating, and productionizing LLM-based AutoRater frameworks or automated regression testing.
Experience designing agentic tooling and integrating agents with commerce platforms.
Experience in conversational AI architectures, agent orchestration harnesses, and retrieval-augmented generation (RAG) for knowledge libraries.
Track record of technical leadership partnering with Product Management, Data Science, and UX to build Ground Truth datasets and performance-tracking dashboards.

Responsibilities

Manage and mentor a Software Engineering team, owning the roadmap and architecture for applied AI, low-latency scoring, and automated evaluation.
Lead Knowledge Library and merchant control engineering, building advanced RAG integrations, including PDF ingestion and auto-learning, for custom agent behavior.
Oversee autonomous execution tooling and Universal Commerce Protocol (UCP) integration to enable active user journeys such as catalog lookups, universal carts, and reservations.
Advocate prompt tuning and loss diagnosis while scaling LLM-based AutoRater frameworks to reduce hallucinations and measure conversation factuality.
Partner with Product, Data Science, UX, and core Google platform teams to define Ground Truth datasets, investigative dashboards, and model upgrade standards.