Agentic AI Architect-Anthropic

This position is listed on behalf of a partner company, who manages all applications and next steps. Our partner is looking for an Agentic AI Architect-Anthropic based in Germany. As an Agentic AI Architect, you will shape and deliver enterprise-grade AI solutions that combine generative AI, machine learning, data, cloud technologies, and intelligent automation. You will design scalable architectures that connect Claude and other AI technologies with enterprise knowledge, applications, APIs, workflows, and mission-critical systems. The role offers significant technical ownership, from discovery and architecture through deployment, evaluation, governance, and optimization. You will work closely with client stakeholders, consultants, data engineers, AI/ML specialists, and platform teams to turn complex business needs into practical solutions. A strong focus will be placed on agentic workflows, RAG, tool use, structured outputs, responsible AI, and human oversight. This is an opportunity to work at the forefront of enterprise AI while helping organizations adopt powerful technologies securely and responsibly. Travel to clients and conferences may be required as needed. Accountabilities Architect, develop, deploy, and maintain scalable AI, machine learning, generative AI, and agentic AI solutions for enterprise use cases. Define end-to-end architectures covering data ingestion and preparation, model selection, orchestration, APIs, integrations, user experiences, monitoring, security, and governance. Partner with business and technical stakeholders to identify high-value AI opportunities, translate requirements into technical designs, and establish delivery roadmaps. Lead technical discovery sessions, architecture workshops, design reviews, proof-of-concepts, demonstrations, and solution delivery activities. Design and implement LLM-powered applications using Claude, Anthropic APIs, and other appropriate LLM platforms. Apply prompt and context engineering techniques, including instruction design, few-shot examples, structured inputs and outputs, response constraints, and long-context management. Architect secure RAG solutions using enterprise documents, knowledge bases, databases, and other approved information sources, including ingestion, chunking, embeddings, retrieval, reranking, citations, and response generation. Build agentic systems capable of reasoning over enterprise context, using authorized tools, executing multi-step tasks, and coordinating workflows across enterprise applications. Define agent roles, task boundaries, permissions, memory and context strategies, approval gates, fallback mechanisms, and escalation paths. Integrate AI agents and workflows with ServiceNow, enterprise APIs, cloud services, databases, collaboration platforms, and operational systems. Implement human-in-the-loop controls and safeguards for sensitive, high-impact, low-confidence, or exception-based actions. Develop machine learning and NLP solutions for predictive analytics, classification, clustering, forecasting, anomaly detection, recommendation, document intelligence, summarization, and intelligent automation. Build data-processing, feature-engineering, ETL/ELT, and model pipelines that support reliable training, deployment, monitoring, and data quality. Design and implement AI solutions across AWS, Microsoft Azure, and/or Google Cloud, using services such as SageMaker, Lambda, S3, and Vertex AI where appropriate. Develop supporting APIs, microservices, automation components, and integrations required to operationalize AI solutions. Work with technologies such as Apache Spark, Snowflake, MySQL, PostgreSQL, MongoDB, and comparable data platforms. Establish AI evaluation frameworks, test suites, representative datasets, regression testing, observability, monitoring, tracing, alerting, and feedback loops. Measure and optimize AI solutions across accuracy, relevance, groundedness, safety, task completion, latency, cost, reliability, and use