Large Language Model Architect

Introduction: Experienced senior architect role responsible for designing and delivering end-to-end enterprise LLM and generative-AI platform architectures, with a particular emphasis on Snowflake-native AI application design. The role is full-time, based in Chennai, and requires extensive experience in software, data and AI architecture to translate business strategy into secure, scalable AI systems. Responsibilities: - Own technical architecture for AI systems spanning classical ML, generative AI, LLM applications, RAG, agentic workflows and enterprise AI platform integration. - Translate product goals into architecture blueprints, non-functional requirements and implementation roadmaps; lead stakeholder workshops to align on feasibility, scope and delivery dependencies. - Define Snowflake-native AI application architectures (Cortex-based agent, RAG, document intelligence, analytics) within Snowflake security perimeter; specify Snowpark and Streamlit patterns and governance (RBAC, masking, lineage, monitoring, cost controls). - Architect model- and tool-agnostic multi-agent systems including orchestration, tool usage, agent memory, context management, control-plane patterns and reusable service abstractions. - Design end-to-end data/context pipelines: ingestion, preprocessing, synchronization, chunking, embeddings, vector search, knowledge graphs and semantic retrieval to support reliable RAG. - Define evaluation frameworks for accuracy, relevance, faithfulness, latency, cost, safety and operational reliability; establish AI security, governance and observability controls (guardrails, prompt-injection defenses, PII protection, audit logging, tracing). - Maintain architecture decision records, diagrams, design specs, integration patterns and reusable reference assets. Required qualifications and experience: - Bachelor's degree or equivalent in Computer Science, Computer Engineering, Data Science, AI/ML or related discipline; 15 years full-time education. - Minimum ~12 years of overall experience; 10+ years in software/data/AI engineering or technology architecture; 5+ years designing and deploying enterprise-grade AI/cloud data solutions; 2+ years in agentic AI/LLM/generative AI architecture; 4+ years coding in Python and experience with APIs, distributed systems and cloud-native patterns; 4+ years in ML/deep learning/NLP/data engineering or AI product delivery. - Demonstrated solution/technology architecture experience in industry contexts (banking, insurance, retail, healthcare, travel, logistics or telecom). Preferred skills and tools: - Hands-on Snowflake Cortex capabilities (Cortex Agents/Search/Analyst/AI Functions), Snowpark, Streamlit, Dynamic Tables, Tasks, Streams, Snowflake ML and Snowflake security features. - Strong knowledge of LLM architecture patterns (RAG, embeddings, vector DBs, prompt engineering, model routing, fine-tuning/adaptation, function calling, agent orchestration). - Experience with CI/CD, IAC, automated testing, model evaluation, MLOps/LLMOps, monitoring and production release governance. - Good to have: SnowPro certifications, exposure to LangChain, LangGraph, LlamaIndex, Haystack, Semantic Kernel, MLflow, FastAPI, Docker, Kubernetes, dbt, responsible AI practices, red-teaming, HITL review and GenAI FinOps. Location & employment: - Role listed for Chennai. Full-time employment. No salary or explicit remote/hybrid working mode stated in the posting.