AI Architect
Description
Job Description - Al Solution Architect
Role: Al Solution Architect - Agentic AI; Experience: 12-15+ years overall experience, with 5+ years in solution architecture, enterprise architecture or Al Agentic solution development; Technology Focus: Generative Al, Agentic Al, Microsoft Azure, MCP, Python,
CICD
Role Overview: Al Solution Architect should be able to lead the design and architecture of enterprise-grade Generative Al and Agentic Al solutions. The architect will be responsible for defining scalable, secure and reusable Al architecture patterns, integrating Large Language Models (LLMs) with enterprise applications, and enabling intelligent workflow automation across business functions. The role involves working closely with business stakeholders, engineering teams, data architects, infrastructure teams and enterprise security teams to translate business requirements into production-ready Al solutions. A strong understanding of enterprise architecture, cloud-native technologies, distributed systems, Al/ML concepts and modern agentic frameworks is essential.
Key Responsibilities
1. Al & Agentic Architecture
- Define end-to-end architecture for enterprise-grade Agentic Al and Generative Al solutions.
- Design multi-agent architectures incorporating autonomous agents, orchestrators, tools, skills, workflows and context management.
- Establish reusable architecture patterns and frameworks for building, deploying and managing Al agents.
- Design agent orchestration, inter-agent communication, task delegation, state management and human-in-the-loop capabilities.
- Define strategies for agent memory, conversation context, session management and knowledge retrieval.
- Evaluate and recommend appropriate agentic.frameworks such as LangGraph, LangChain and other enterprise agent frameworks.
2. LLM & Generative Al Integration
- Architect integrations with enterprise LLM platforms and models through secured
APls and enterprise Al gateways. - Design prompt engineering, structured output, function/tool calling and model selection strategies.
- Implement Retrieval-Augmented Generation (RAG), embeddings, vector search and enterprise knowledge integration where required.
- Define strategies for LLM evaluation, response quality, hallucination mitigation, guardrails and responsible Al.
- Optimize model performance, latency, scalability and token consumption.
3. Enterprise Integration & MCP Architecture
- Design integration between Al agents and enterprise systems through REST APIS, microservices and Model Context Protocol (MCP).
- • Architect MCP servers and clients for secure exposure and consumption of enterprise tools, services and data sources.
- Integrate Al solutions with enterprise platforms such as CRM, underwriting systems, document repositories and other business applications.
- Define API contracts, service communication patterns, authentication mechanisms and integration standards.
- Design asynchronous communication and event-driven workflows using messaging platforms and queues.
4. Cloud Architecture & Infrastructure
- Define cloud-native architecture using Azure Container Services, Azure Service Bus,
- Azure API Management, Azure Storage and other relevant Azure services.
- Architect containerized application deployments using Docker, Kubernetes and OpenShift.
- Establish CI/CD, RevSecORs and automated deployment practices for Al applications.
5. Security, Governance & Compliance
- Design enterprise-grade security architecture for Al solutions.
- Implement authentication and authorization using Microsoft Entra ID (Azure AD), service principals, managed identities and Auth 2.0.
- Define secure credential and secret management using Azure Key Vault or equivalent enterprise solutions.
- Establish secure communication between agents, MCP servers, APIs, LLM* platforms and enterprise backend systems.
- Address data privacy, encryption, access control, auditability and regulatory compliance.
- Define Al governance, responsible Al, model risk management and data protection practices.
6. Solution Delivery & Tehnical Leadership
- Translate business requirements into architecture blueprints, solution designs and technical specifications.
- Lead architecture reviews, technology evaluations and design discussions with stakeholders.
- Guide engineering teams on implementation, coding standards, integration patterns and deployment strategies.
- Define architecture decision records (ADRs), technical standards, reusable components and reference architectures.
- Support proof of concepts, technical validations, performance testing and production readiness assessments.
- Collaborate with enterprise architects, infrastructure, cybersecurity, data and application development teams.
Preferred Domain Experience: Experience in insurance or other regulated enterprise environments, Understanding of commercial insurance and underwriting processes, including submission intake, risk assessment, quote generation, binding and policy issuance. Experience implementing Al-driven workflow automation and straight-through processing (STP). Experience integrating Al solutions with enterprise platforms such as Salesforce, CRM, document management and policy administration systems.
Understanding of multi-region enterprise deployments, data residency and regional compliance requirements.
Key Success Criteria: The architect will be expected to deliver scalable, secure, reusable and production-ready Al architectures that accelerate enterprise Al adoption, enable intelligent business process automation, improve operational efficiency and support future
Al transformation initiatives.
Senior Al engineer
- A Senior Al Engineer is responsible for designing, developing, and implementing Al solutions to solve complex business problems. Key responsibilities include:
- Designing and implementing Al systems: Building ad training machio learning models and integrating them into applications.
- Testing and optimizing models: Evaluating model performance, tuning hyperparameters, and ensuring reliability in production.
- Collaborating with cross-functional teams: Working closely with domain experts and developers to translate business needs into technical requirements.
- Managing Al infrastructure: Overseeing cloud computing platforms and ensuring efficient model deployment and scaling.
- Documenting processes: Keeping detailed records of models, workflows
Required Technical SKIls
words
Skill Area | Required Expertise |
Generative Al | LLMs, Prompt Engineering, RAG, Function Calling |
Agentic Al | Multi-Agent Systems, Agent Orchestration, LangGraph, LangChain |
MCP | Model Context Protocol, MCP Servers, MCP Clients, Tool Integration |
Cloud | Microsoft Azure, Cloud-Native Architecture |
Programming | Python |
Architecture | Microservices, Distributed Systems, Event-Driven Architecture |
Integration | API Management, Auth 2.0, Enterprise System Integration |
Data | PostgreSQL, Cosmos DB, ADLS, Vector Databases |
Messaging | Azure Service Bus, Queues, Topics, Event-Driven Workflows |
Containers | Docker, Kubernetes, OpenShift |
Security | Azure AD / Entra ID, Managed Identity, Service Principals, Key Vault |
DevOps | GitHub, CI/CD, Azure DevOps, RexSecORs, Infrastructure as Code |
Skills
- Agentic AI
- AI
- API
- Authentication
- Automation
- Azure
- Azure AD
- Azure DevOps
- Azure Service Bus
- CI/CD
- Cloud
- Cloud Native
- Cosmos Db
- CRM
- Cybersecurity
- DevOps
- Distributed Systems
- Docker
- Embeddings
- Entra ID
- Event Driven Architecture
- GitHub
- Infrastructure as Code
- Kubernetes
- LangChain
- LangGraph
- LLM
- Machine Learning
- MCP
- Microservices
- Model Deployment
- OpenShift
- PostgreSQL
- Prompt Engineering
- Python
- RAG
- Regulatory Compliance
- REST
- Risk Assessment
- Salesforce
- System Integration
- Vault
- Vector Databases
- Vector Search
Sign in to read and join the discussion.