Overview

: AI/ML Solution Architect Agentic AI & Generative AI

Responsibilities

• 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.
• 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.
• 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.
• 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.
• Enterprise Integration & Cloud Architecture
• Design Python-based AI services, APIs, microservices, and integration layers using frameworks such as FastAPI.

Qualifications

• 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.
• Mandatory Technical Requirements
• SkillExpected Expertise
• Preferred Skills
• Hands-on experience with Docker, Kubernetes, and container orchestration.
• Knowledge of PostgreSQL, Redis, graph databases, and knowledge graphs.
• Familiarity with AI security, guardrails, responsible AI, and enterprise AI governance.
• 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.