Agentic Commerce Architect

Accenture Song accelerates growth and value for our clients by combining creativity, technology, and data-driven intelligence. Within Song, our Commerce & AI practice helps organisations design, build, and scale modern digital commerce ecosystems - embedding Generative AI and Agentic AI into commerce to enable real-time personalisation, autonomous merchandising, and intelligent customer engagement at scale.

About the role...

We are entering an exciting era of agentic commerce where AI agents, conversational interfaces, and intelligent systems are reshaping how people discover, evaluate, and buy online. We are looking for technically strong architects who can help design and deliver the systems that make this real - working at the intersection of enterprise commerce platforms and emerging agentic AI technologies.

You are an experienced technical architect with hands-on delivery experience across enterprise commerce platforms and a growing, demonstrable specialism in Generative AI and agentic systems. You are comfortable owning the technical design of a solution workstream, making well-reasoned build vs. buy decisions, and translating complex architecture into clear recommendations for both engineering teams and senior client stakeholders.

You do not need to have designed enterprise-scale agentic systems end-to-end, but you must have meaningful hands-on experience with the core components: LLM-powered applications, RAG architectures, agent frameworks, and enterprise platform integration. You are intellectually curious, technically rigorous, and energised by working at the frontier of what is currently possible in commerce AI.

Key Responsibilities

• Owning the technical architecture of agentic commerce solution workstreams within enterprise delivery programmes, designing system components, integration patterns, and ensuring the overall technical approach is sound, scalable, and aligned to client constraints.

• Designing and documenting agentic AI solution architectures — including multi-agent patterns, LLM-powered applications, retrieval-augmented generation (RAG) systems, and orchestration pipelines applied to commerce use cases such as personalisation, conversational commerce, intelligent search, and merchandising optimisation

• Architecting the integration between agentic AI systems and enterprise platforms — including OMS, PIM, CRM, payments, and fulfilment — via APIs, event-driven patterns, and emerging agentic protocols

• Contributing to build vs. buy decisions across LLM platforms, orchestration frameworks, vector stores, and relevant commerce platforms — bringing a well-reasoned technical point of view grounded in delivery experience

• Designing the knowledge and data architecture that underpins agent intelligence — including embedding pipelines, vector retrieval, semantic search, and structured product data — ensuring agent outputs are accurate, grounded, and commercially reliable

• Embedding AI governance, observability, and responsible AI principles into architecture design — including audit logging, human-in-the-loop escalation points, and performance monitoring — as first-class concerns rather than afterthoughts

• Providing technical leadership to delivery teams — guiding engineers, reviewing technical decisions, and maintaining architectural coherence across workstreams throughout the delivery lifecycle

• Leading technical workshops and architecture review sessions with client stakeholders — presenting options with clear trade-offs, effort implications, and business outcomes in terms that resonate with both technical and non-technical audiences

• Supporting the development of Accenture Song's agentic commerce architecture capability — including contributing to reference architectures, technical accelerators, and points of view that advance our delivery standards

Required Experience

• ~7 years of professional experience in technical architecture, solution engineering, or software engineering — with meaningful delivery experience across enterprise platforms and a growing specialism in AI and agentic systems

• Experience leading technical design sessions and presenting architecture options to senior client or stakeholder audiences — able to communicate complex technical decisions with clarity and confidence

• Hands-on experience delivering on at least one major enterprise commerce platform — such as Salesforce Commerce Cloud, Adobe Commerce, Commercetools, SAP Commerce, or a composable / MACH-aligned stack — with a solid understanding of how these platforms integrate with surrounding enterprise systems

• Demonstrable, hands-on experience with Generative AI and agentic AI systems — including LLM-powered application design, retrieval-augmented generation (RAG) architectures, and at least one agentic AI framework (e.g.LangGraph, LangChain, LlamaIndex, CrewAI)

• Experience designing and building data and context pipelines for AI systems — including embedding pipelines, vector databases, and semantic search applied to structured commerce data

• Familiarity with emerging agentic commerce protocols — such as MCP, A2A, or ACP — and how they are being adopted to connect AI agents with enterprise commerce systems and product data

• Working knowledge of cloud platforms (AWS, GCP, or Azure) for AI deployment, including familiarity with cloud-native services and CI/CD-based delivery practices

• Understanding of AI governance and responsible AI principles — including observability, tracing, auditability, and how to build guardrails into production AI systems

Preferred Experience

• Experience contributing to technical proposals, reference architectures, or delivery accelerators in a consulting or professional services context

• Hands-on experience with hyperscaler agentic platforms — such as Google Vertex AI Agent Builder, AWS Bedrock Agents, or Azure AI Foundry — and how they are applied in commerce delivery contexts

• Hands-on experience with AgentOps or LLMOps practices — including production monitoring, model evaluation, tracing, and continuous improvement frameworks for deployed AI systems

• Familiarity with GEO (Generative Engine Optimisation) and AI-native search — including how product data must be structured and exposed to remain visible and accurate in LLM-mediated discovery environments

• Exposure to composable commerce architecture patterns — including MACH principles, headless commerce, and API-first platform design — and how these enable more flexible agentic AI integration