AI Solution Architect

Responsibilities Translate business problems into solution designs and technical specs — choosing the right approach (including simpler non-AI options) for the problem. Architect end-to-end AI/ML solutions: data access and pipelines, model integration or development, retrieval/knowledge layers, agents, and application integration. Applied ML (e.g. vision, NLP, OCR) and GenAI both fall in scope. Define reusable architecture patterns, standards, and the review process that keeps team-built and BU-built solutions aligned and production-ready. Set and steward data-access, PII, and security guardrails for AI applications, in partnership with governance. Evaluate and recommend AI/ML technologies; own build-vs-buy-vs-fine-tune-vs-enable decisions, and establish evaluation, observability, and cost practices for production. Provide technical leadership across the solution lifecycle and partner with Data Engineering, Data Science, Data Platform, Data Product, BU builders / AI Champions, and Platform/IT. Requirements Experience as a Solution Architect and/or System Analyst delivering production AI/ML or data products (retail a plus). Breadth across applied ML and GenAI application architecture — LLMs, RAG, agents, prompt engineering, and evaluation — with the judgment to mix pre-trained, fine-tuned, and custom models as the problem warrants. Solid data engineering and cloud platform foundations (GCP is a plus). Hands-on with AI/ML frameworks and APIs — agent/orchestration frameworks, vector stores, and model APIs/MCP. Working knowledge of data governance, PII handling, security, and responsible-AI principles — and how to operationalize them.