: AI/ML Solution Architect Agentic AI & Generative AI

Senior Leadership / Architecture Role

10+ Years Experience

Position Details

ParameterDetails

Job TitleAI/ML Solution Architect Agentic AI / Generative AI

Experience10+ Years

Open Positions1

Location - Noida

Work Mode - Work From Office (WFO)

Employment Type Full time

Primary Technologies Python, LangGraph, MCP, A2A, LiteLLM, LLMs

Job Summary

We are looking for an experienced AI/ML Solution Architect with 10+ years of experience in AI/ML, Generative AI, and enterprise solution architecture to lead the design and implementation of scalable, secure, and production-ready AI solutions.

The ideal candidate will have strong hands-on expertise in Agentic AI, LangGraph, MCP (Model Context Protocol), A2A (Agent-to-Agent), DAG-based workflow orchestration, and LiteLLM, along with proven experience architecting enterprise-grade LLM-powered applications and multi-agent systems.

This role requires a combination of strategic architecture leadership, deep technical expertise, and hands-on implementation capabilities. The architect will be responsible for defining AI architecture, establishing engineering standards, evaluating technology choices, and guiding development teams in delivering reliable, scalable, and cost-effective AI solutions.

The candidate will work closely with business stakeholders, engineering teams, and leadership to translate business requirements into enterprise AI solutions, from architecture and proof of concept through production deployment and ongoing optimization.

Key Responsibilities

1. AI/ML Solution Architecture

- Define and own end-to-end architecture for enterprise AI/ML, Generative AI, and Agentic AI solutions.
- Design scalable, modular, secure, and cloud-native architectures for LLM-powered applications and intelligent automation platforms.
- Establish architecture principles, design patterns, technology standards, and best practices for enterprise AI development.
- Evaluate and select appropriate foundation models, LLM frameworks, orchestration tools, vector databases, and cloud services based on business and technical requirements.
- Translate complex business use cases into technical architecture, solution blueprints, and implementation roadmaps.
- Lead architecture reviews, technical design discussions, and proof-of-concept initiatives.

2. Agentic AI & LangGraph Architecture

- Architect and implement stateful, multi-step, and autonomous AI agent workflows using LangGraph.
- Design advanced multi-agent architectures supporting collaboration, task delegation, planning, reasoning, and decision-making.
- Define agent state management, memory strategies, checkpointing, persistence, human-in-the-loop workflows, and recovery mechanisms.
- Design agent orchestration patterns for complex enterprise use cases involving multiple tools, APIs, data sources, and LLMs.
- Establish reusable frameworks and components for developing and deploying Agentic AI applications.
- Guide engineering teams in implementing production-grade LangGraph solutions with appropriate testing, error handling, and performance optimization.

3. MCP, A2A & Workflow Orchestration

- Design and implement MCP-based architectures for connecting AI agents with enterprise applications, APIs, databases, and external tools.
- Define standards for MCP server development, tool discovery, authentication, authorization, and secure tool execution.
- Architect A2A communication and interoperability between autonomous agents and distributed AI services.
- Design DAG-based workflows for deterministic execution, dependency management, orchestration, and fault recovery.
- Establish patterns for integrating LangGraph with MCP, A2A,



and workflow orchestration frameworks.
- Evaluate and implement orchestration technologies to support reliable, scalable, and maintainable AI workflows.

4. LLM Engineering & Model Strategy

- Define enterprise LLM integration and model strategy across OpenAI, Azure OpenAI, Anthropic, Gemini, and other foundation model providers.
- Architect LiteLLM-based model abstraction, centralized access, provider routing, fallback mechanisms, and multi-model integrations.
- Design RAG and advanced RAG architectures using embeddings, vector databases, hybrid search, reranking, and knowledge retrieval.
- Establish standards for prompt engineering, structured outputs, function calling, tool use, and agent memory.
- Evaluate models based on accuracy, latency, throughput, token consumption, cost, and business requirements.
- Implement strategies for model selection, model evaluation, versioning, and continuous improvement.

5. Enterprise Integration & Cloud Architecture

- Design Python-based AI services, APIs, microservices, and integration layers using frameworks such as FastAPI.
- Architect integrations with enterprise systems, databases, SaaS applications, and third-party platforms.
- Define deployment architectures across AWS, Azure, and GCP, including cloud-native AI infrastructure.
- Establish containerization, Kubernetes, CI/CD, infrastructure automation, and deployment standards.
- Design scalable data pipelines, retrieval systems, and secure data access mechanisms for AI applications.
- Ensure interoperability between AI platforms, enterprise systems, and existing technology ecosystems.

6. Security, Governance & Responsible AI

- Define AI security architecture covering data privacy, access control, identity management, and secure model access.
- Implement guardrails for prompt injection, data leakage, unsafe tool execution, and unauthorized agent actions.
- Establish governance frameworks for model usage, data handling, auditability, and enterprise AI compliance requirements.
- Design human approval and escalation mechanisms for sensitive or high-impact AI workflows.
- Define observability and audit standards for agent decisions, tool calls, model interactions, and workflow execution.
- Collaborate with security and compliance teams to ensure enterprise AI solutions meet organizational standards.

7. Performance, Reliability & Production Excellence

- Establish architecture standards for AI application monitoring, logging, tracing, and end-to-end observability.
- Define strategies for reducing LLM inference costs, latency, and unnecessary token consumption.
- Design fault-tolerant architectures with retries, timeouts, fallback models, circuit breakers, and recovery mechanisms.
- Implement evaluation frameworks for measuring response quality, groundedness, retrieval accuracy, and agent task completion.
- Guide teams in resolving production issues and improving application reliability and scalability.
- Drive continuous architecture optimization based on production performance and business outcomes.

8. Technical Leadership & Stakeholder Management

- Provide technical leadership and mentorship to AI/ML engineers, senior developers, and solution teams.
- Lead architecture workshops, requirement discussions, technical presentations, and solution demonstrations.
- Collaborate with business stakeholders, product teams,



and engineering leadership to define AI adoption strategies.
- Review code, technical designs, architecture documents, and implementation approaches.
- Own technical delivery from solution design and estimation through deployment and production support.
- Identify emerging AI technologies and assess their applicability to enterprise use cases.

Required Skills & Qualifications

Mandatory Technical Requirements

SkillExpected Expertise

Overall Experience10+ years in software engineering, AI/ML, data science, or related technology domainsAI/ML & GenAIStrong experience architecting and delivering enterprise AI/ML and Generative AI solutionsLangGraphExtensive hands-on, project-based experience designing and implementing production-grade LangGraph systemsAgentic AIExpertise in autonomous agents, multi-agent systems, planning, reasoning, and orchestrationMCPStrong experience implementing MCP servers, tools, and enterprise integrationsA2AExperience designing agent-to-agent communication and multi-agent interoperabilityDAG WorkflowsStrong understanding of DAG-based orchestration, workflow dependencies, and fault-tolerant executionLiteLLMHands-on experience with LLM abstraction, model routing, provider integrations, and fallback strategiesPythonAdvanced Python programming and AI backend developmentLLM & RAGStrong expertise in LLMs, RAG, advanced RAG, embeddings, vector databases, and prompt engineeringCloudExperience architecting and deploying solutions on AWS, Azure, or GCPArchitectureProven experience with distributed systems, microservices, APIs, scalability, and enterprise integration

Preferred Skills

- Experience with LangChain, LlamaIndex, CrewAI, or AutoGen.
- Experience with FastAPI, REST APIs, and event-driven architectures.
- Hands-on experience with Docker, Kubernetes, and container orchestration.
- Experience with vector databases such as Pinecone, Weaviate, Milvus, Qdrant, or FAISS.
- Knowledge of PostgreSQL, Redis, graph databases, and knowledge graphs.
- Experience with LLMOps, MLOps, CI/CD, model evaluation, and AI observability frameworks.
- Familiarity with AI security, guardrails, responsible AI, and enterprise AI governance.
- Experience in designing reusable AI platforms, shared services, and enterprise AI accelerators.

Candidate Screening Mandatory Criteria

Critical Hiring Requirements

- Candidates must have 10+ years of overall relevant technology experience, with substantial AI/ML and Generative AI architecture exposure.
- LangGraph experience must be hands-on and demonstrated through real project implementation, preferably in production environments.
- Candidates with only theoretical knowledge, certifications, training, or basic POCs in LangGraph should not be considered sufficient.
- Practical experience with MCP, multi-agent architectures, A2A, DAG workflows, and LiteLLM is strongly expected.
- The candidate must demonstrate both architecture ownership and the ability to go hands-on with Python and AI solution development.
- Must be comfortable working from the Noida office.

Core Technology Stack Primary keywords for sourcing and screening

Python | AI/ML | Generative AI | Agentic AI | Solution Architecture | LangGraph | MCP | Model Context Protocol | A2A | Multi-Agent Systems | DAG | LiteLLM | LLMs | RAG | Advanced RAG | Vector Databases | FastAPI | AWS | Azure | GCP | LLMOps | AI Governance

Suggested Naukri / LinkedIn search title: AI/ML Solution Architect | GenAI Architect | Agentic AI Architect | Principal AI Engineer | Generative AI Architect

Contact -

Call / What's App Resume - (phone hidden)

Email - [email protected]

MANAGER GLOBAL TALENT ACQUISITION

MOPTRA INFOTECH PVT LTD

📌 Solution Architect AI/ML Agentic AI & Generative AI (Delhi)
🏢 Moptra Infotech
📍 Delhi