Vertex AI Engineer

Job Title: Vertex AIJob Location: Zaandam, NetherlandsJob Type : PermanentJob Mode: Hybrid - 3 days onsite and 2 days remote in a weekRole / Job Purpose: The Vertex AI Engineer is responsible for designing, developing, deploying, and supporting AI/GenAI solutions on Google Cloud Vertex AI. The role focuses on building scalable AI applications, RAG solutions, AI agents, MLOps pipelines, model deployment, monitoring, governance, and responsible AI implementation.Responsibilities:AI Application Development· Design and build AI/GenAI solutions using Vertex AI.· Develop enterprise AI assistants, copilots, and AI agents.· Build Retrieval Augmented Generation (RAG) solutions.· Integrate Gemini and other foundation models into enterprise applications.· Implement prompt engineering and model optimization.Model Engineering· Train, fine-tune, evaluate, and deploy ML/LLM models.· Utilize Vertex AI Model Garden and Gemini models.· Manage model lifecycle from experimentation to production.· Optimize model performance, latency, and cost.MLOps & Platform Engineering· Build and maintain Vertex AI Pipelines.· Implement CI/CD for AI models.· Manage model registry, feature stores, and artifacts.· Automate deployment and monitoring processes.AI Operations & Governance· Monitor model performance and drift.· Implement observability, logging, and alerting.· Apply Responsible AI and security controls.· Manage API quotas, IAM roles, and access controls.· Ensure compliance with privacy and enterprise governance requirements.· Build and manage Grounding configurations — Google Search grounding and custom data source grounding via Vertex AI Agent Builder for factual, up-to-date responses.· Design multi-modal AI applications leveraging Gemini 2.0+ capabilities — text, image, audio, video, and document inputs within a single unified model call.· Implement AI agent evaluation frameworks — using Vertex AI Evaluation Service (AutoSIA, BLEU, ROUGE, custom metrics) to assess agent quality before production deployment.· Configure and manage Vertex AI Endpoints with traffic splitting, canary deployments, and autoscaling for production-grade model serving SLAs.· Use Vertex AI Model Evaluation and Vertex AI Experiments for A/B testing between model versions (champion/challenger) with statistically significant evaluation datasets.· Implement AI safety controls — configure Vertex AI Safety Filters (harm categories, thresholds), input/output sanitization, and PII detection for enterprise-grade Gemini deployments.· Apply AI governance frameworks — EU AI Act compliance considerations, model cards, data lineage documentation, and bias evaluation using Vertex Explainable AI (SHAP values, feature attributions).· Manage Generative AI token cost governance — implement token budgeting, prompt length optimization, caching strategies (Vertex AI Context Caching), and per-application quota enforcement.