Hands‑on n8n experience: Production workflows or similar low‑code workflow engines; focus on reliability, idempotency, and failure recovery.
Strong SQL & BigQuery skills: Data modeling, performant queries, and production‑grade pipelines in GCP/BigQuery (Composer/Airflow, Dataflow is a plus).
API integration expertise: Experience with authentication (OAuth/service accounts), pagination, retries/backoff, schema evolution.
GCP IAM experience: Designing and implementing least‑privilege access for GCP resources (projects, service accounts, BigQuery datasets/tables, secrets, service‑to‑service auth).
LLM production usage: Experience with OpenAI, Vertex AI, or similar; prompt design, evaluation, cost/latency tradeoffs.
Software engineering fundamentals: Skilled in Python or TypeScript/Node; comfortable with testing, code reviews, CI/CD.
Collaboration & communication skills and ability to work cross‑functionally and influence decisions.
Nice to Have
Experience with Vertex AI, Dataform, or other GCP‑native ML tooling.
Platform/enablement team experience building tools for other engineers.
Roles & Responsibilities
Design, build, and operate n8n workflows that orchestrate GenAI use cases end‑to‑end, from event triggers to evaluation, monitoring, and rollout.
Build and document custom connectors and integrations for n8n.
Integrate with internal and external REST/GraphQL APIs, including authentication, rate limiting, and robust error handling patterns.
Build and optimize BigQuery data models and pipelines that feed, monitor, and evaluate GenAI workflows (prompt logs, evaluation datasets, feature stores, cost dashboards).
Implement and maintain GCP IAM and permissions models (projects, service accounts, dataset/table access, secrets) to keep GenAI workflows secure and auditable.
Partner with AI Platforms and product teams to integrate LLM Gateway or other LLM provider APIs, including prompt design, safety/guardrail patterns, and offline evaluation loops.
Define and instrument observability (structured logs, metrics, dashboards, alerts) for n8n + LLM workflows using Wayfair‑standard tooling (e.g., DataDog, logging pipelines).
Collaborate with engineering, data science, and product stakeholders to translate fuzzy GenAI ideas into well‑scoped, testable workflows with clear success metrics.
Contribute to documentation, runbooks, and reusable templates so non‑expert teams can safely adopt GenAI Accelerator patterns.
