Ai Architect

About the Role We are looking for an experienced AI / ML Architect to lead the design and implementation of advanced Generative AI and RAG (Retrieval-Augmented Generation) solutions. The role combines hands-on architecture design, pre-sales engagement, and technical leadership across enterprise AI initiatives. You will drive solutioning around LLMs, knowledge retrieval, and MCP-based multi-agent architectures, helping customers unlock business value from AI responsibly and at scale. Key Responsibilities Architect and deliver enterprise-grade AI / ML & Generative AI solutions, including RAG pipelines, LLM integrations, and intelligent agents. Engage in pre-sales activities : collaborate with business development, present technical solutions, estimate effort, and support proposals / PoCs for prospects. Design knowledge retrieval layers using vector databases (FAISS, Pinecone, Milvus, Chroma, Weaviate). Develop document ingestion, embedding, and context-retrieval pipelines for unstructured and structured data. Architect and manage MCP (Model Context Protocol) servers for secure context exchange, multi-model orchestration, and agent-to-agent collaboration. Define LLMOps / MLOps best practices – CI / CD for models, prompt versioning, monitoring, and automated evaluation. Collaborate with pre-sales and business teams to shape AI solution proposals, PoCs, and client demos. Lead AI innovation initiatives and mentor technical teams on GenAI, RAG, and MCP frameworks. Ensure data privacy, compliance, and responsible AI across all deployments Work closely with ITS, TIC team to provide mentorship and guidance to AI developers Required Skills & Experience 12–15 years of overall experience with 5–7 years in AI / ML and 3+ years in Generative AI / LLM architecture. Strong hands-on experience with RAG pipelines, vector search, and semantic retrieval. Proven experience integrating LLMs (OpenAI, Claude, Gemini, Mistral, etc.) using frameworks such as LangChain, LlamaIndex, or PromptFlow. Deep understanding of MCP servers – configuration, context routing, memory management, and protocol-based interoperability. Strong programming skills in Python, and familiarity with containerization (Docker, Kubernetes) and cloud AI services (Azure OpenAI, AWS Bedrock, GCP Vertex AI). Expertise in MLOps / LLMOps tools (MLflow, KubeFlow, LangSmith, Weights & Biases). Solid grounding in data engineering, pipelines, and orchestration tools (Airflow, Prefect). Excellent communication, client engagement, and technical presentation skills Proven track record of practice building or leadership in emerging technology domains. Preferred / Good to Have Experience integrating MCP servers with LangChain agents or OpenAI’s MCP ecosystem for scalable orchestration. Knowledge of RAG evaluation frameworks (RAGAS, TruLens) and hallucination-reduction techniques. Experience with enterprise data connectors (SharePoint, Confluence, SQL / NoSQL, APIs). Familiarity with Knowledge Graphs and hybrid retrieval (symbolic + neural). Exposure to agentic frameworks like LangGraph, CrewAI, or AutoGen. Proven experience building or scaling AI / GenAI Centers of Excellence (CoEs). Why Join Dexian Be part of a growing Enterprise AI practice shaping next-generation intelligent systems. Opportunity to architect cutting-edge GenAI and RAG solutions for global clients. Collaborative and innovation-driven culture with deep technical mentorship.