Director - AI Product Architect

We are looking for an AI Product Architect – Digital Finance to lead the architecture and technical evolution of an AI-enabled Digital Finance product platform.

The role will sit at the intersection of Finance, Product Engineering, Data, Generative AI and Agentic AI. The architect will translate finance business capabilities into scalable product architecture and ensure AI capabilities are secure, governed, explainable, reusable and production-ready.

The Digital Finance architecture includes role-based finance experiences, finance intelligence, AI workers/agents, governed knowledge and controls, and trusted finance data products. 

This is a product architecture role, not just an AI/LLM development role. The successful candidate should be capable of taking an AI use case from business problem through architecture, design, engineering patterns, controls and production deployment.

Responsibilities

Key Responsibilities

1. Product Architecture

• Own end-to-end architecture for the Digital Finance AI product ecosystem.

• Translate product vision and finance use cases into scalable, modular architecture.

• Define reusable architectural patterns rather than point solutions for individual finance use cases.

• Architect capabilities spanning user experience, orchestration, AI processing, data, integrations, security and platform services.

• Establish architecture standards, design principles, APIs and reusable product components.

2. Generative AI & Agentic AI Architecture

• Design enterprise-grade AI agents and multi-agent workflows for finance use cases.

• Define patterns for agent orchestration, tool/API invocation, memory/context, RAG, semantic/vector search, structured data retrieval and human-in-the-loop workflows.

• Establish an LLM gateway / model abstraction approach so the product can work with approved foundation models without tightly coupling product capabilities to an individual model.

• Design guardrails, grounding, validation, observability and fallback mechanisms.

• Establish evaluation approaches covering quality, accuracy, latency, cost and user feedback.

These areas directly align with the current Digital Finance technical direction, which identifies LLM gateway, agent orchestration, vector storage, APIs, monitoring and audit storage as core components. 

3. Finance Intelligence Architect AI capabilities supporting areas such as:

• Record-to-Report / Period Close

• Accounts Receivable and collections

• Accounts Payable

• Invoice and contract intelligence

• Cash and liquidity

• Working capital

• Planning and forecasting

• Reconciliation and anomaly detection

• Tax, audit and controls

• CFO insights and management narratives

The existing product architecture specifically identifies capabilities including Close Intelligence, reconciliation, GL anomaly detection, contract-to-invoice validation, DSO/collections intelligence, cash forecasting, working-capital optimization and finance narratives. 

4. Data & Knowledge Architecture

• Define how AI consumes governed structured and unstructured finance information.

• Design reusable finance data products and semantic/knowledge layers.

• Architect RAG and retrieval patterns across policies, contracts, invoices, SOPs and other finance knowledge.

• Establish grounding and lineage from source information through AI-generated responses.

• Work closely with data engineering teams on lakehouse/data-product architecture.

The current technical requirements use Microsoft Dynamics, Azure Databricks, Delta Lake, Power BI, Unity Catalog, APIs and document stores as components of the Digital Finance landscape. 

5. AI Governance, Trust & Security

• Define architecture for responsible and trustworthy AI.

• Implement guardrails against hallucination, inappropriate actions and unsupported recommendations.

• Design model/prompt versioning, evaluation and monitoring.

• Ensure traceability of AI recommendations to supporting evidence.

• Implement identity, RBAC, sensitive-data protection and audit logging.

• Design human approval controls for high-risk finance actions.

The internal requirements explicitly call for human approval for high-risk actions, as well as model governance, prompt governance, audit trails and data lineage. 

6. Product Engineering Leadership

• Work closely with Product Managers, Finance SMEs, UX, AI/ML engineers, data engineers, application engineers and security teams.

• Review solution designs and provide architecture guidance to engineering squads.

• Drive technical PoCs into reusable, production-quality product capabilities.

• Make architecture decisions balancing business value, accuracy, scalability, performance, security, cost and maintainability.

• Mentor engineering teams on modern AI architecture and product-engineering practices.

Required Experience

10+ years of technology/product engineering experience, with meaningful hands-on experience in modern AI/data architecture

Candidates should demonstrate:

• Strong enterprise/application architecture background.

• Hands-on understanding of Generative AI, LLM and Agentic AI architectures.

• Experience designing production AI applications, not only PoCs.

• Strong understanding of: 

  • LLM orchestration

  • RAG and semantic search

  • Vector databases

  • Embeddings

  • Agent/tool calling

  • Prompt/context engineering

  • LLM evaluation

  • AI observability

  • Guardrails and responsible AI

  • API/microservices architecture

• Experience with cloud-native architecture.

• Strong data architecture understanding, preferably including lakehouse and governed data-product patterns.

• Experience integrating AI systems with enterprise platforms and workflows.

Preferred Technology Experience

AI / GenAI

• Azure OpenAI / OpenAI APIs or comparable enterprise LLM platforms

• Agentic AI frameworks and orchestration patterns

• RAG and vector search

• Model evaluation and monitoring

• Prompt and model lifecycle management

Data & Cloud

• Azure

• Azure Databricks

• Delta Lake

• Unity Catalog

• SQL / Python

• Data APIs and semantic layers

Enterprise & Integration

• REST/API architecture

• Event-driven and microservices patterns

• Microsoft Dynamics 365

• Power Platform / Power Automate

• Power BI

• Enterprise identity and RBAC

Several of these technologies already appear in the Digital Finance technical architecture, particularly Dynamics, Databricks/Delta Lake, Unity Catalog, Power BI, APIs and workflow integration. 

Finance Domain Experience

Deep finance expertise should be preferred rather than mandatory if you want to maximize the AI architecture talent pool.

Exposure to one or more of the following would be valuable:

• Record-to-Report

• Order/Lead-to-Cash

• Procure-to-Pay

• Accounts Receivable / Accounts Payable

• Working Capital

• Treasury / Cash Management

• Financial Planning & Analysis

• Financial Close and Reconciliation

• Audit, Controls and Compliance

Qualifications

Bachelor’s or higher degree in Computer Science or a related discipline; or equivalent