Job Description

We are seeking a Director level executive for Lead AI Architect with proven expertise in Large Language Models (LLMs), Agentic AI, LangChain, LangGraph, embeddings, vector databases, and semantic search. The role demands cloud expertise (Azure or GCP), application architecture knowledge, and experience in designing and delivering enterprise-grade AI applications using LLMs and agentic workflows.


The ideal candidate blends technical excellence with a product mindset, capable of applying design patterns in LangChain and LangGraph, as well as integration architecture patterns, to deliver scalable, reusable, and business-aligned AI solutions.




Key Responsibilities


Overall experience: 15+ years of IT experience and strong experience with Data architectures

AI Product Strategy & Delivery Enterprise AI & Application Architecture LLM & Agentic AI Development Embeddings, Vector Databases & RAG Cloud & MLOps Leadership & Hands-On Execution Lead the end-to-end AI product lifecycle -- ideation, prototyping, MVP, and enterprise-grade rollout. Ensure all AI solutions are aligned with business KPIs, adoption metrics, and ROI goals. Collaborate with product, business, and architecture teams to shape AI-enabled solutions. Architect scalable enterprise AI applications using LLMs, embeddings, and agentic workflows. Apply integration design patterns (e.g., API Gateway, event-driven, CQRS, pub-sub, orchestration vs. choreography) for seamless enterprise system integration. Ensure security, compliance, and resilience in all deployed solutions. Design and implement LLM-powered applications (OpenAI, Gemini, Claude, LLaMA). Use LangChain design patterns (e.g., Retrieval Chain, Conversational Chain, Sequential Chains, Tool-Using Agents) to create modular and reusable workflows. Apply LangGraph design patterns for multi-agent orchestration, tool invocation, and branching logic in complex workflows. Implement agentic AI architectures with Model Context Protocol (MCP) for adaptive and autonomous decision-making. Build and optimize embedding pipelines (OpenAI Embeddings, SentenceTransformers, custom models). Architect semantic search and RAG pipelines with vector databases (FAISS, Weaviate, Pinecone, Milvus). Ensure low-latency, scalable retrieval for enterprise knowledge applications. Deploy AI workloads on Azure AI (Azure OpenAI, Azure Cognitive Search, Azure ML, Foundry) or Google Cloud AI (Vertex AI, Gemini, PaLM). Lead MLOps practices for automated training, CI/CD, monitoring, and retraining. Apply cloud-native design patterns (e.g., sidecar, adapter, strangler fig, service mesh) for AI deployment. Be a hands-on leader, actively coding prototypes and leading architecture reviews. Mentor teams in LangChain/LangGraph design patterns, embeddings, RAG, and AI solution design. Drive best practices in AI development, integration design, and scalable deployment.



Primary skills8+ years in AI/ML development, with 3+ years in enterprise LLM/agentic AI solutions. Hands-on expertise with LangChain, LangGraph, and design patterns for workflow orchestration. Deep understanding of agentic AI architectures, embeddings, RAG pipelines, and semantic search. Strong Python development skills and experience integrating AI models with enterprise APIs and microservices. Experience with Azure AI or GCP AI services for enterprise-scale deployments. Knowledge of application and integration design patterns for cloud-native architectures. Proficiency with vector databases (FAISS, Pinecone, Weaviate, Milvus). Experience with MLOps pipelines, CI/CD, and Kubernetes. Strong stakeholder management, communication, and leadership skills

Ref. code
330995-en_GB
Posted on
26 Nov 2025
Experience level
Executives
Contract type
Permanent
Location
Gurgaon
Business unit
I and D Global Business Line
Brand
Capgemini
Professional communities
Architecture