AI/ML Solution Architect - Agentic Automation (Managed Services) - Consulting

Job description

Job Summary EY Consulting is hiring an AI/ML Solution Architect to design, build, and industrialize agentic automation solutions for Managed Services across HR, Finance, Procurement, Supply Chain, Risk, Tax, and other enterprise functions. The role combines hands-on engineering with solution architecture, bringing together Agentic AI, GenAI, workflow orchestration, enterprise integration, cloud-native platforms, and managed-service operating models to improve productivity, quality, cycle time, compliance, and user experience. The role is a practical technologist who can whiteboard an architecture with executives, prototype an agentic workflow with engineers, and guide teams through secure, reliable, and cost-efficient production delivery on Microsoft Azure, AWS, or Google Cloud Platform. The role requires strong experience with LLMs, RAG, agents, integration patterns, automation platforms, MLOps/LLMOps, and enterprise-grade governance. Key Responsibilities 1) Solution Architecture for Agentic Managed Services Design end-to-end agentic automation architectures for managed-service processes such as hire-to-retire, record-to-report, procure-to-pay, source-to-contract, order-to-cash, service desk, knowledge operations, and compliance operations. Translate business outcomes into solution blueprints, capability maps, technical roadmaps, non-functional requirements, success measures, and implementation backlogs. Define reusable reference architectures for AI agents, RAG, workflow orchestration, human-in-the-loop review, exception handling, audit trails, and enterprise knowledge management. Balance build, buy, and partner options across hyperscalers, AI platforms, automation tools, enterprise SaaS, and EY assets. 2) Hands-on Engineering and Prototyping Build working PoCs, MVPs, accelerators, and production components using Python, TypeScript, APIs, microservices, event-driven patterns, and cloud-native services. Implement RAG pipelines, tool-calling agents, orchestration graphs, evaluation harnesses, prompt and policy controls, and observability dashboards. Develop integrations with enterprise systems such as SAP, Oracle, Workday, ServiceNow, Coupa, Ariba, Microsoft 365, Dynamics, Salesforce, and document management platforms. Guide engineering teams on coding standards, CI/CD, test automation, infrastructure as code, release management, and operational runbooks. 3) Cloud and Platform Architecture Architect secure, scalable solutions on one or more hyperscaler stacks: Microsoft Azure, AWS, or Google Cloud Platform. Use native AI, data, integration, identity, security, and observability capabilities, including Azure AI Foundry/Azure OpenAI, AWS Bedrock/SageMaker, Google Vertex AI/Gemini, cloud data platforms, serverless services, container platforms, and managed Kubernetes. Design hybrid and regulated deployment patterns covering private networking, identity federation, secrets management, encryption, data residency, model risk, and compliant logging. Define cost-control mechanisms including model routing, caching, batching, token governance, scaling policies, FinOps dashboards, and usage analytics. 4) Agentic Automation and Process Transformation Design agentic patterns such as planning, routing, delegation, tool use, memory, reflection, approval workflows, and multi-agent collaboration for enterprise operations. Apply process-mining, workflow, and task-automation concepts to redesign managed-service processes before automating them. Create human-in-the-loop controls for sensitive steps such as payment approvals, employee actions, vendor changes, reconciliations, policy exceptions, and regulatory submissions. Define measurable operating outcomes, including automation rate, exception rate, first-time-right quality, handling time, SLA compliance, leakage reduction, and cost-to-serve improvement. 5) Reliability, Security, Risk, and Governance Establish LLMOps and MLOps practices for model/prompt versioning, evaluation, guardrails, monitoring, rollback, incident response, and quality assurance. Embed AI governance controls for responsible AI, data privacy, access control, auditability, explainability, model risk, and regulatory compliance. Implement production observability across logs, traces, metrics, user feedback, groundedness, hallucination risk, tool execution, cost per request, and service-level performance. Lead design reviews, threat modeling, architecture assurance, performance tuning, and post-implementation optimization. 6) Client Advisory, Pursuits, and Delivery Leadership Partner with client executives, managed-service leaders, function owners, CIO/CTO teams, and ecosystem partners to shape AI-led transformation opportunities. Lead discovery workshops, value framing, solution estimation, PoVs/PoCs, business cases, acceptance criteria, and transition plans from prototype to managed operations. Coach cross-functional teams across EY, client, and partner organizations, including architects, engineers, data scientists, process SMEs, security teams, and operations leads. Create reusable assets, architecture playbooks, demo journeys, and delivery patterns for ASEAN priority industries and service lines. Required Qualifications 10+ years of experience across AI/ML, solution architecture, platform engineering, data engineering, enterprise automation, or cloud-native application delivery. Hands-on experience delivering production AI/ML, GenAI, RAG, conversational assistant, or agentic automation solutions at enterprise scale. Strong proficiency in at least one major cloud stack: Microsoft Azure, AWS, or Google Cloud Platform, including AI services, data services, identity/security, networking, and deployment patterns. Practical software engineering capability in Python and one or more of TypeScript, Java, C#, or Go; strong understanding of APIs, microservices, integration design, and testing strategies. Experience with LLMOps/MLOps practices such as evaluation, prompt/version management, model registry, monitoring, CI/CD, guardrails, and release governance. Knowledge of enterprise workflow and automation patterns across HR, Finance, Procurement, Supply Chain, or shared-services operations. Strong understanding of security, privacy, responsible AI, data residency, access control, audit logging, and model risk considerations. Client-facing consulting experience, including structured problem solving, executive communication, workshop facilitation, solution shaping, and delivery leadership. Preferred Qualifications Experience with managed-services or shared-services operating models, including process transition, service catalogues, SLAs, runbooks, knowledge management, and continuous improvement. Hands-on experience with agent frameworks and orchestration tools such as LangGraph, Semantic Kernel, AutoGen, CrewAI, OpenAI Assistants, or equivalent frameworks. Experience with automation and workflow platforms such as Microsoft Power Platform, UiPath, Automation Anywhere, ServiceNow, Camunda, Temporal, Airflow, or cloud-native workflow services. Experience integrating with ERP, HCM, procurement, and service-management platforms such as SAP, Oracle, Workday, Coupa, Ariba, ServiceNow, Dynamics, and Microsoft 365. Familiarity with vector databases and search technologies such as Azure AI Search, pgvector, Pinecone, Milvus, Redis, OpenSearch, Elasticsearch, or BigQuery/Vertex AI search patterns. Relevant cloud or architecture certifications such as Azure Solutions Architect, Azure AI Engineer, AWS Solutions Architect, AWS Machine Learning, Google Professional Cloud Architect, or Google Professional Machine Learning Engineer. Experience in regulated industries such as financial services, public sector, health, energy, or cross-border environments with data-sovereignty requirements. Technical Skills GenAI, Agents, and RAG LLM application design, RAG, embeddings, chunking, hybrid search, knowledge graphs, tool calling, function calling, planning, routing, memory, and multi-agent orchestration. Evaluation and safety: groundedness, relevancy, hallucination risk, regression testing, red teaming, policy enforcement, PII controls, content moderation, and prompt hardening. Cloud and Platform Microsoft: Azure OpenAI, Azure AI Foundry, Azure AI Search, Azure Machine Learning, AKS, Functions, Logic Apps, Event Grid, Key Vault, Entra ID, Purview, Monitor, Fabric or Synapse. AWS: Amazon Bedrock, SageMaker, Lambda, Step Functions, EKS, ECS, Glue, OpenSearch, DynamoDB, S3, IAM, KMS, CloudWatch, EventBridge, and data lake patterns. Google Cloud: Vertex AI, Gemini, GKE, Cloud Run, Cloud Functions, BigQuery, Dataflow, Pub/Sub, Apigee, Cloud IAM, Secret Manager, Cloud Logging, and Cloud Monitoring. Engineering, Integration, and Operations Python, TypeScript, REST/GraphQL APIs, event-driven architecture, containers, Kubernetes, Terraform/Bicep/CloudFormation, GitHub Actions/Azure DevOps/GitLab CI, and automated testing. Enterprise integration patterns for ERP, HCM, procurement, ITSM, CRM, document repositories, email/chat channels, OCR/document intelligence, and workflow engines. Observability and operations: OpenTelemetry, Prometheus/Grafana, cloud-native monitoring, Langfuse/Arize/WhyLabs or equivalents, incident management, SLAs/SLOs, and runbooks. Soft Skills Strong executive presence with the ability to simplify complex AI, automation, and architecture topics for business and technology leaders. Hands-on leadership style: comfortable moving from strategy and architecture into code, prototyping, troubleshooting, and delivery problem solving. Outcome orientation with a strong focus on measurable value, adoption, operational resilience, and continuous improvement. Ability to lead cross-border, cross-functional teams and manage ambiguity across business, technology, security, risk, and operations stakeholders. Travel Requirements Singapore-based with ASEAN travel as needed for client delivery, pursuits, workshops, and regional leadership engagements. Relocation support may be available for the right candidate.