Role : Sr. AI Solutions Architect GCP & Agentic Systems

Role Summary

This is a senior, full-stack AI practitioner role for someone who can operate as both builder and architect lead. The successful candidate will define AI architecture, build and deploy AI/ML solutions, establish scalable data and MLOps foundations, and act as the go-to technical expert on client engagements. You will be USDMs technical voice in all working sessions expected to hold design conversations with Google AI engineers, not just facilitate them.

This role combines AI Architect, AI/ML Engineer, Data Engineer, MLOps Lead, and AI Risk/Controls capability into one person. The right candidate moves fluidly between strategy, design, delivery, and governance. Life sciences or pharma experience is strongly preferred.

Experience & Background :

- 7+ years across AI/ML engineering, data engineering, solution architecture, or technical product delivery

- Proven hands-on experience delivering AI/ML or GenAI solutions from concept to production show us what youve shipped

- Direct production experience with Vertex AI agentic pipelines not prototypes

- Active GCP Professional certification this is a hard requirement, not a preference

- Experience working in a consulting or client-facing delivery model

- Strong communicator across technical and non-technical audiences you can run a design session with engineers and a stakeholder briefing in the same day

- Bachelors or Masters degree in Computer Science, Data Science, Engineering, or related field

Technical Skills :

GCP & Data Infrastructure :

- Vertex AI: model deployment, pipelines, online/batch prediction, Model Registry

- Gemini API integration (Pro, Flash, or Ultra)

- Vertex AI Agent Builder / Google Agent Development Kit (ADK)

- Cloud Dataflow and Apache Beam for stream and batch pipeline engineering

- Airflow / Cloud Composer for pipeline orchestration

- BigQuery: data modeling, ML integration, and analytics pipelines

- Cloud Storage, Pub/Sub, Cloud Functions, and Cloud Run

- GCP IAM, VPC Service Controls, CMEK, service account governance, and audit logging

AI / LLM Development :

- RAG pipeline design: ingestion, chunking, embedding, vector search, and grounded retrieval

- Multi-agent orchestration: tool use, delegation, memory, state management, and failure handling

- Prompt engineering for production: system instructions, few-shot, chain-of-thought, output formatting

- Model evaluation: consistency testing, hallucination assessment, and output quality measurement

- LLM observability: drift detection, latency monitoring, error rates, and revalidation triggers

- LangChain, LangGraph, or equivalent orchestration frameworks

Engineering Fundamentals :

- Python primary language; production-grade, reviewable code

- FHIR R4 API integration (R5 a strong advantage)

- REST API design and enterprise system integration patterns

- Docker, Cloud Run, or GKE for containerized AI service deployment

- Terraform or GCP Deployment Manager for infrastructure-as-code

- CI/CD pipelines using Cloud Build, GitHub Actions, or equivalent

- SQL and data modeling for structured and unstructured AI data pipelines

Strong Advantage :

- Life sciences, pharma, CDMO, or GxP regulated environment experience

- Google Professional Services or large GSI GCP practice background

- FHIR R5 depth sufficient to hold design conversations with FHIR engineers

- Familiarity with large-scale enterprise data environments and data governance frameworks

- Experience with AI governance, risk frameworks (ISO 42001, NIST AI RMF), or EU AI Act risk classification

- Validation or qualification documentation experience for AI/ML systems (URS, test protocols, RTM)