Principal Engineer, Agentic Applications
Gap Inc. · SF - 2 Folsom; 4440 Rosewood Drive
What the posting says
Each fact was read from this posting and checked:Statedin the posting, quotedConflictsposted two ways, both shownNot mentionedwe looked; it is not there
- Employment typeStated
- Full time
Not mentioned in this posting
We read the whole posting. These are not in it. Not a no: ask the recruiter, in your own words.
- PayNot mentioned
- The posting does not say
- Work arrangementNot mentioned
- The posting does not say
- ShiftNot mentioned
- The posting does not say
- ExperienceNot mentioned
- The posting does not say
- EducationNot mentioned
- The posting does not say
- Visa sponsorshipNot mentioned
- The posting does not say
Requirements, each with its sentence from the posting
These are the employer’s words. The index does not judge whether you meet them.
- No requirement has been itemized for this posting yet. The employer’s words are in the full posting below.
Read the full posting as the employer wrote it
About the Role
In this role, you will bridge creative design, product creation, and the global supply chain for Gap Inc. This is not a research role — you will build and ship AI/ML-backed multi-agent workflows that turn structured & unstructured data, mood boards, tech packs, fabric specs, vendor collaboration, etc. into live, sometimes autonomous, production decisions.As a senior, hands-on engineering leader, you will direct the technical standards of the agentic systems that allow our business to operate at AI-native speed and scale. You will drive this transformation while owning the agentic application and AI-ready data designs, and co-owning the reference architecture and agent framework for our product-to-market journey.
What You'll Do
Own Technical Standards. Set and enforce engineering standards, contracts, and integration patterns — including interoperability protocols such as MCP/A2A — for agentic solutions across digital apparel design tools, PLM systems, vendor management, procurement, compliance, logistics, etc. Set the engineering definition-of-done, establishing system quality through evaluation gates (LLM-as-judge, golden datasets) rather than subjective opinion
Architect Agent-Ready Data. Shape the data strategy for how structured and unstructured design and sourcing assets are ingested, embedded, structured, and exposed as reliable, reusable AI-ready data products — that provide features for sciences and ontology + context for multi-agent loops
Co-Own the Agentic Reference Architecture. With peer AI/ML platform teams, design the agent harness (including trust frameworks), core orchestrator pattern, semantic/business-context layers, and the tool contracts for the ML and data platforms while being model agnostic
Lead Complex Product-to-Market Agent Workflows. Build and deploy multi-agent systems that orchestrate tool-use and multi-step reasoning across Design, Development, and Sourcing — auto-generating BOMs, negotiating RFPs, allocating materials, etc.
Partner on Guardrails & Governance. Drive FinOps including cost-aware routing across Vertex Model Garden based on latency, cost, and capability, backed by real data. Collaborate closely with Trust, Security, and Governance teams to ensure agents ship safe, grounded, entitlement-aware, and gated against non-deterministic failures
Elevate the Engineering Bar. Mentor data and agent engineers across delivery pods and vendor-augmented teams by building alongside them, not just reviewing PRs; represent agentic engineering in architecture reviews with senior technology leadership
Who You Are
12+ years, hands-on. In software/data/ML engineering, with recent experience building and shipping production AI/ML systems — ideally LLM-based agents or multi-agent orchestration, not just classical ML pipelines. You're still writing and reviewing code by choice, not solely reviewing architecture diagrams
Staff/Principal/Architect track record. At a large, complex enterprise, you've set technical standards that other engineering teams were expected to follow, not just proposed them
Production agent and cloud AI experience. Comfort with orchestration patterns (e.g., LangGraph, custom orchestrators, or equivalent), and hands-on depth with at least one major cloud AI stack (GCP/Vertex preferred). You've built or owned LLM-as-judge pipelines, golden datasets, or comparable quality gates for a production AI system, not just discussed them conceptually
Comfortable with conflict. Across engineering, sciences, platform, senior leadership, and security, you stand behind core design principles and rigor
The posting text is reproduced as the employer wrote it; the index adds nothing to it.