AI Architect Agentic AI

AI Solution Architect GenAI & Agentic AI

Job Title: AI Solution Architect GenAI & Agentic AI

Experience: 10+ Years

Relevant AI/ML/GenAI Architecture Experience: 4+ Years

Location: Bengaluru, Delhi/NCR, Hyderabad, Chennai, Pune, Mumbai.

Employment Type: Full-Time (Hybrid)

Industry: IT Services / Consulting / AI Transformation

Job Overview

We are looking for an experienced AI Solution Architect GenAI & Agentic AI to lead the architecture, design, governance, and implementation of enterprise-scale Generative AI and Agentic AI solutions.

The ideal candidate will have strong experience in AI/ML architecture, Generative AI, Large Language Models (LLMs), Agentic AI, RAG, AI orchestration, cloud AI platforms, LLMOps, AI governance, and enterprise solution architecture.

The candidate will be responsible for translating business requirements into scalable and secure AI architectures, driving AI platform adoption, establishing responsible AI practices, and enabling production-grade GenAI and multi-agent applications.

Experience with Google Vertex AI or Azure AI Foundry, LangGraph, LangChain, LangSmith, Python, RAG, Vector Databases, LLMOps, Kubernetes, Docker, CI/CD, and AI governance is highly desirable.

---

Key Responsibilities

1. AI Solution Architecture

• Design and architect end-to-end Generative AI and Agentic AI solutions aligned with enterprise business objectives.

• Define scalable, secure, resilient, and reusable AI reference architectures and solution patterns.

• Lead architecture discussions, design reviews, technology evaluations, and architecture governance.

• Define integration patterns between AI platforms, enterprise applications, APIs, data platforms, and business systems.

• Evaluate emerging AI technologies, models, frameworks, and platforms.

• Drive enterprise adoption of AI-first engineering and AI-assisted software development practices.

• Provide technical leadership and guidance to engineering and AI/ML teams.

• Translate business and functional requirements into technical AI architecture and implementation roadmaps.

2. Generative AI & LLM Architecture

• Design enterprise solutions using Large Language Models (LLMs) such as Azure OpenAI, OpenAI models, Gemini, and other foundation models.

• Define strategies for model selection, model deployment, fine-tuning, prompt optimization, and inference.

• Architect solutions for accuracy, scalability, latency, cost optimization, security, and explainability.

• Define LLM integration patterns with enterprise applications and APIs.

• Establish standards for prompt engineering, prompt lifecycle management, testing, and evaluation.

• Design LLM evaluation frameworks and quality metrics.

3. Agentic AI & Multi-Agent Architecture

• Architect Agentic AI and multi-agent systems using frameworks such as LangGraph, LangChain, and LangSmith.

• Design deterministic and non-deterministic AI workflows.

• Define agent routing, orchestration, tool calling, reasoning, state management, and workflow execution patterns.

• Design multi-agent collaboration and delegation mechanisms.

• Implement memory management and context management strategies.

• Design Human-in-the-Loop (HITL) approval, escalation, and exception-handling mechanisms.

• Establish guardrails to control agent behavior and prevent unintended actions.

• Evaluate and implement frameworks such as Semantic Kernel, CrewAI, and AutoGen where appropriate.

4. RAG, Knowledge & Context Architecture

• Design enterprise-grade Retrieval Augmented Generation (RAG) architectures.

• Architect knowledge stores, context stores, vector databases, and semantic search solutions.

• Define embedding strategies and vector search architectures.

• Design document ingestion, chunking, indexing, retrieval, reranking, and grounding approaches.

• Implement strategies for context engineering and LLM memory management.

• Design semantic search and knowledge retrieval frameworks.

• Evaluate advanced architectures such as GraphRAG and Knowledge Graphs.

• Ensure enterprise knowledge is securely and accurately grounded within AI applications.

5. AI Platform Architecture

Architect and govern AI solutions using one or more of the following platforms:

• Google Vertex AI

• Azure AI Foundry

• Azure OpenAI

- Gemini

- OpenAI ecosystem

Define and implement enterprise AI platform capabilities including:

• Model hosting and deployment

• Model gateway

• Prompt management

• Model lifecycle management

• AI observability

• AI safety and guardrails

• LLMOps

• Model monitoring

• Usage and cost tracking

• AI application security

6. LLMOps / MLOps & AI Engineering

- Define enterprise LLMOps and MLOps practices.

- Establish CI/CD pipelines for AI and GenAI applications.

- Design model versioning, prompt versioning, evaluation, deployment, and rollback strategies.

- Implement monitoring for model performance, latency, cost, quality, and reliability.

- Integrate AI solutions with GitHub, Azure DevOps, Kubernetes, Docker, and enterprise DevOps platforms.

- Define production-readiness standards for AI applications.

7. Vibe Coding & AI-Assisted Development

• Lead adoption of AI-assisted software development and Vibe Coding practices.

• Define enterprise patterns for AI-assisted coding and automated code generation.

• Implement agent-assisted software development and testing.

• Establish AI-based code review and quality-assurance practices.

• Define governance standards for AI-generated code.

• Establish metrics to measure developer productivity and quality improvements through AI-assisted development.

• Promote responsible use of coding assistants and autonomous development agents.

8. AI Governance, Risk & Responsible AI

• Define and implement enterprise AI governance and Responsible AI frameworks.

• Conduct AI risk assessments and model risk reviews.

• Establish controls for:

• Data privacy

• Data security

• Regulatory compliance

• AI auditability

• Hallucination mitigation

• Prompt injection protection

• Data leakage prevention

• Model monitoring

• Human oversight

• Explainability and transparency

• Define security and compliance requirements for enterprise AI applications.

• Establish policies for responsible development, deployment, and usage of GenAI.

• Ensure AI solutions comply with organizational policies and applicable regulatory requirements.

9. AI Security

• Design secure architectures for enterprise GenAI and Agentic AI applications.

• Implement controls against prompt injection, jailbreaks, data leakage, unauthorized tool execution, and malicious AI inputs.

• Define authentication, authorization, identity, access control, and API security mechanisms.

• Collaborate with cybersecurity teams on AI threat modelling and AI security assessments.

• Experience with AI red teaming and adversarial testing is an advantage.

10. Stakeholder Management & Technical Leadership

• Work closely with business stakeholders, enterprise architects, engineering teams, product owners, security teams, and senior leadership.

• Present AI architecture, technical strategies, risks, and recommendations to executive stakeholders.

• Provide AI strategy and solution consulting for enterprise transformation initiatives.

• Mentor architects, engineers, and AI/ML teams.

• Establish reusable AI architecture patterns, accelerators, frameworks, and best practices.

• Lead technology evaluations, PoCs, and transition of successful AI solutions into production.

---

Mandatory Skills & Experience

AI / GenAI Platforms

• Strong experience with Azure AI Foundry OR Google Vertex AI.

• Experience with Azure OpenAI, Gemini, OpenAI ecosystem, or equivalent LLM platforms.

• Strong understanding of LLM architecture and AI application development.

• Experience deploying production-grade AI/GenAI solutions.

• Knowledge of LLMOps / MLOps practices.

Agentic AI

• Strong hands-on/architectural experience with LangGraph.

• Experience with LangChain and LangSmith.

• Understanding of multi-agent architecture and agent orchestration.

• Knowledge of deterministic and non-deterministic workflows.

• Experience with tool calling, routing, memory, state management, and Human-in-the-Loop workflows.

• Semantic Kernel experience is preferred.

LLM Engineering

• Prompt Engineering and Prompt Optimization

• LLM evaluation and benchmarking

• RAG architecture

• Embeddings

• Vector databases

• Semantic search

• Context engineering

• LLM memory management

• Model selection and optimization

Cloud & Software Engineering

• Strong Python programming experience.

• APIs and Microservices

• Kubernetes

• Docker

• CI/CD

• GitHub / Azure DevOps

• Cloud architecture and enterprise integration

AI Governance & Security

• Responsible AI

• AI Governance Frameworks

• AI Risk Management

• Data Privacy and Security

• Regulatory Compliance

• AI Auditability

• AI Safety and Guardrails

• Prompt Injection Protection

• Model Monitoring

---

Good to Have

• CrewAI / AutoGen

• Databricks AI Platform

• MLflow

• PromptFlow

• Neo4j / Knowledge Graphs

• GraphRAG

• AI FinOps / GenAI Cost Optimization

• AI Security and Red Teaming

• BFSI / Banking / Financial Services / Insurance domain experience

• Enterprise Architecture experience

• Consulting / Solution Consulting experience

---

Preferred Certifications

• Microsoft Certified: Azure AI Engineer Associate

• Microsoft Azure AI / AI Foundry certifications

• Google Professional Machine Learning Engineer

• Google Vertex AI certifications

• Responsible AI / AI Governance certifications

• TOGAF / Enterprise Architecture certification

---

Key Competencies

• Enterprise AI Architecture

• GenAI & Agentic AI Strategy

• Solution Architecture

• AI Strategy & Advisory

• Enterprise Transformation

• Executive Stakeholder Management

• Solution Consulting

• Technology Leadership

• Innovation & Emerging Technologies

• Risk-Based Decision Making

• AI Governance

• Team Mentoring

• Architecture Governance