Job Summary:

The Principal Engineer operates at the intersection of deep technical execution and broad organizational influence. This role will resolve the hardest architectural questions, set the AI-native engineering bar, and ensure that iHerb's platform decisions compound over time rather than accumulate debt. Reporting to the VP of Engineering this role collaborates closely with peer Principals, Engineering Managers, and peer VPs across the organization.

This role owns the end-to-end technical strategy for iHerb's AI Platform and GenAI product experiences. This spans the user-facing AI layer (Wellness Agent, LLM-driven recommendations, RAG over catalog and reviews, generative content) and the shared AI infrastructure (RAG pipelines, evals framework, MLOps, guardrails, observability) that makes AI features production-grade and repeatable across the engineering organization.

Job Expectations:

Define and own the AI platform architecture: retrieval infrastructure, model lifecycle, evals framework, guardrails, and GenAI product feature design.
Lead the build of the shared AI platform layer consumed by all AI product features and reusable by internal business teams.
Hands-on contributor: build production AI systems, write proofs of concept, and validate architecture through working software.
Set and enforce technical standards for the AI Platform team; drive architecture reviews and model quality reviews.
Coordinate with the Personalization team to define clear boundaries between the GenAI product layer and existing ML personalization infrastructure.
Contribute AI platform-specific patterns and lessons into iHerb's shared AI-driven SDLC golden path.
Drive the hardest cross-cutting technical decisions across multiple teams and shared platform services.
Establish and evolve iHerb's AI-driven SDLC golden path: shared standards, Claude Code skills, guardrails, and automation patterns.
Lead complex multi-team technical efforts by coordinating architecture reviews, aligning peer Principals and EMs, and resolving competing approaches.
Mentor and raise the technical bar across the engineering organization through code review, architecture review, and direct coaching of senior engineers.
Represent engineering in cross-functional conversations with product, data science, security, and infrastructure.
Feed architectural decisions into the shared knowledge base so institutional knowledge compounds across the organization.

The duties and responsibilities described above may provide only a partial description of this position. This is not an exhaustive list of all aspects of the job. Other duties and responsibilities not outlined in this document may be added as necessary or desirable, with or without notice.

Knowledge, Skills and Abilities:

Required:

AI-driven SDLC (required): demonstrated use of AI-assisted development tools such as Claude Code, GitHub Copilot, or Cursor to ship production systems. Can articulate workflow changes, quality tradeoffs, and guardrail strategies.
Architecture at scale: experience designing and evolving large-scale distributed systems across multiple teams and years: APIs, data pipelines, event-driven architectures, or high-traffic platforms.
Cross-org technical leadership: track record of driving architectural standards, technical roadmaps, or platform initiatives that span multiple teams or organizations.
Engineering quality mindset: deeply held opinions on code quality, observability, CI/CD, test automation, and maintaining velocity without accumulating hidden debt.
Communication and influence: able to write clear architecture documents, present to technical and non-technical audiences, and build consensus without formal authority.
Experience working in distributed teams across the US, China, and Latin America.
Production experience designing and operating AI platform systems: RAG pipelines, vector search, embedding infrastructure, or retrieval-augmented applications at scale.
Hands-on experience with LLM evaluation: building eval frameworks, defining quality metrics, and iterating on model and prompt quality using data.
MLOps experience: model deployment, monitoring, lifecycle management, and cost governance in a production environment.
Experience building and operating agentic systems using frameworks such as LangChain, LlamaIndex, or equivalent.
GenAI in production is required; prototype or research-only experience is not sufficient at this level.

Principal Engineer – AI Platform at iHerb