Purpose
The AI Solution Architect – GSS leads high-priority AI initiatives, leveraging AWS Bedrock, LangGraph and advanced agentic AI technologies to solve complex business challenges for data scientists, AI agents, and business users. Acting as the bridge between functional domain experts and technical engineering teams, this role translates business needs into robust, scalable, and well-governed AI-powered solutions.
This position is accountable for the end-to-end design, deployment, and user experience of AI and analytics products supporting GSS (IT Support, Finance, HR, Indirect Procurement, Legal, Facility Management, etc.). The AI Solution Architect owns the full AI solution landscape delivered by IT&D teams, including architectural design, proof-of-concept implementation, AI model integration, and management of development, testing, and production environments.
As a key enabler in the GSS AI journey, the AI Solution Architect ensures AI solutions are interoperable, ethically governed, performant, and built on the right platforms. The role works closely with the Central Data & Analytics Architecture team and AI Center of Excellence to align functional and enterprise-wide needs, ensure the right AI technologies and frameworks are in place, and embed Corporate Functions within the organization's AI framework.
Accountabilities
Solution Design for Corporate Functions:
- Design and implement AWS Bedrock-centered AI architectures and agentic patterns to meet the needs of Finance, HR, Procurement, Legal, and other corporate functions.
- Ensure AI solutions meet requirements for intelligent automation, conversational AI, decision support, and predictive analytics, while aligning with enterprise AI governance, security, and responsible AI standards.
AI Agent & Solution Architecture:
- Architect agentic AI solutions using LangGraph and multi-agent frameworks, defining agent orchestration, tool integration, memory management, and human-in-the-loop patterns for corporate function use cases.
- Translate business requirements from corporate stakeholders into scalable, maintainable AI solution designs that leverage foundation models, RAG (Retrieval-Augmented Generation), and agent-based workflows.
AI Model Integration & Platform Build:
- Lead the design of AI solution architectures integrating AWS Bedrock foundation models (Claude, Titan, etc.) with enterprise data sources, vector databases, and knowledge bases.
- Guide AI Engineers and Data Engineers in implementing LangGraph-based agent workflows, prompt engineering strategies, and model evaluation frameworks.
Leadership & Collaboration:
- Lead and mentor AI Engineers, Data Engineers, SMEs, and external consultants working on corporate function AI initiatives.
- Work closely with business analysts, product owners, AI governance teams, and legal/compliance to ensure solutions meet functional, ethical, and regulatory requirements.
Integration & Orchestration:
- Architect agentic integration solutions using LangGraph for multi-step reasoning, tool calling, and external system integration (SAP, Workday, ServiceNow, etc.).
- Collaborate with platform teams to implement CI/CD pipelines for AI agents, including automated testing, model versioning, and deployment automation on AWS.
AI Modernization & Transformation:
- Contribute to AI transformation strategies, migrating legacy automation and analytics solutions to modern agentic AI architectures powered by Bedrock and LangGraph.
- Define patterns for responsible AI deployment, including monitoring, explainability, bias detection, and continuous improvement of AI agents in production.
Professional Experience:
- 15+ years of overall professional experience, including:
- 10+ years in architecting enterprise-scale AI solutions.
- 7+ years enterprise software development, solution engineering, or AI/ML consulting
- 3+ years production experience with Large Language Models and generative AI
- 4+ years hands-on experience with AWS SageMaker (training, deployment, pipelines, feature store)
- 2+ years hands-on experience with Langgraph, LangChain, Bedrock or Agentic AI frameworks
- Proven track record delivering complex AI projects from conception to production
Technical Expertise:
AI Platforms & Cloud:
- Deep expertise in AWS Bedrock (Claude, Titan, Jurassic models) and AWS AI/ML services (SageMaker, Kendra, OpenSearch, Textract, Comprehend).
- Proficient with AWS infrastructure for AI workloads: Lambda, ECS/EKS, S3, DynamoDB, RDS, API Gateway, EventBridge, Step Functions.
- Experience with vector databases (Pinecone, Weaviate, pgvector, OpenSearch Vector Engine) for RAG implementations.
Agentic AI Frameworks & LLM Orchestration:
- Expert in LangGraph for building stateful, multi-agent systems with cyclic workflows, human-in-the-loop patterns, and complex reasoning chains.
- Proficient in LangChain ecosystem (agents, tools, chains, memory, callbacks) and prompt engineering techniques.
- Experience with agent frameworks (Langgraph, CrewAI, Semantic Kernel) and multi-agent orchestration patterns.
Programming & AI Development:
- Proficient in Python for AI solution development, including libraries: boto3, langchain, langgraph, pydantic, FastAPI, streamlit.
- Experience with prompt engineering, few-shot learning, chain-of-thought reasoning, and model evaluation frameworks.
- Skilled in building conversational AI interfaces and integrating with collaboration platforms (Slack, Teams, ServiceNow).
Integration & Tooling:
- Skilled in API integration for AI agents (REST, GraphQL, webhooks) and tool-calling patterns for external system access.
- Experience with CI/CD for AI (GitHub Actions, AWS CodePipeline) including automated testing, model versioning, and deployment strategies.
- Proficient with collaboration platforms (Jira, Confluence, Bitbucket) and AI observability tools (LangSmith, Weights & Biases, MLflow).
AI Architecture Best Practices:
- Ability to design solutions aligned with AWS Well-Architected Framework for AI/ML and Responsible AI principles across security, governance, explainability, bias mitigation, and cost optimization.
- Expertise in RAG architecture patterns: chunking strategies, embedding models, retrieval optimization, and context management.
- Knowledge of AI safety and guardrails: content filtering, PII detection, hallucination mitigation, and model monitoring.
Enterprise Systems & Knowledge Integration:
- Experience integrating AI agents with enterprise systems (SAP ECC/S4, Ariba, Workday, Salesforce, ServiceNow) via APIs and connectors.
- Skilled in knowledge base construction from structured and unstructured corporate data sources (SharePoint, Confluence, databases, PDFs).
- Understanding of enterprise authentication (SSO, OAuth, SAML) and secure API access patterns for AI agents.
Multi-Cloud & AI Platform Awareness:
- Understanding of Azure OpenAI Service, Azure AI Studio, and Google Cloud Vertex AI for hybrid and multi-cloud AI strategies.
- Awareness of alternative LLM platforms (Anthropic Direct, OpenAI, Cohere, Hugging Face) and model selection criteria.
- Knowledge of edge AI deployment patterns and on-premises LLM hosting options for sensitive use cases.
Key Competencies:
- Strategic Thinking & Problem Solving: Ability to analyse and simplify complex problems, articulate trade-offs for informed decision-making, and make appropriate recommendations.
- Collaboration & Influence: Excellence in collaboration, cross-functional expectation management, and influencing skills.
- Business Acumen: Ability to translate business requirements into databases, data warehouses, and data streams.
- Data Management: Experience in creating procedures to ensure data accuracy and accessibility; analysing, planning, and defining data architecture frameworks (including security, reference data, metadata, and master data); and creating and implementing data management processes and procedures.
- Stakeholder Engagement: Proven ability to collaborate with other teams within the organization to devise and implement data strategies, build models, and assess shareholder needs and goals.
- Innovation: Aptitude for researching data acquisition opportunities and developing application programming interfaces (APIs) to retrieve data.
- Communication: Excellent written, verbal, and meeting facilitation skills.