Gen AI Architect

Job Title: Gen AI ArchitectExperience: 8 – 12 YearsWork Mode: Work from Office (WFO)Locations: Chennai | Bangalore | HyderabadPosition Overview We are looking for an experienced GenAI Architect to lead the design and implementation of cutting-edge Generative AI solutions that drive innovation, scalability, and enterprise transformation.This role demands deep expertise in LLM-based architectures, RAG (Retrieval-Augmented Generation) systems, agentic workflow orchestration, and MLOps deployment. The ideal candidate will combine strong hands-on technical skills with architectural vision, ensuring secure, compliant, and high-performance GenAI ecosystems.Key ResponsibilitiesGenerative AI (GenAI) Architecture- Architect and design enterprise-grade LLM-based systems optimized for performance, scalability, and reliability.- Demonstrate strong understanding of vector embeddings, similarity search (cosine/IP/L2), chunking strategies, and reranking.- Build and optimize RAG pipelines — including indexing, metadata, hybrid search, and evaluator frameworks.- Apply prompt engineering for tool use, function calling, and multi-agent planning.- Develop and orchestrate agentic workflows using frameworks such as LangGraph or similar; integrate tools/services via MCP-compatible patterns.- Implement NL2SQL approaches, ensuring SQL safety through schema constraints, sandboxes, and secure API integrations.- Evaluate and balance trade-offs between generic LLMs and fine-tuned models, optimizing for accuracy, latency, and operational cost.- Embed security, compliance, and data governance principles within all AI workflows, including RBAC/ABAC, auditability, and privacy controls.MLOps & Deployment- Manage end-to-end deployment using frameworks such as MLflow, Kubeflow, SageMaker, or Vertex AI.- Implement CI/CD pipelines for AI models, ensuring consistent delivery and robust version control.- Utilize Docker and Kubernetes for model containerization, orchestration, and scalability.- Establish monitoring, tracing, and observability for model performance, latency, and cost metrics.Preferred Experience- Hands-on experience with cloud platforms such as AWS, Azure, GCP, or OCI.- Strong understanding of large-scale distributed systems and big-data frameworks (Spark, Hadoop).- Expertise in retrieval optimization (hybrid lexical + vector search, metadata filtering, rerankers).- Knowledge of model fine-tuning and adapter techniques (LoRA, SFT, DPO) with evaluation best practices.- Experience in Document AI, layout parsing, and schema design for unstructured data.- Familiarity with observability stacks for LLM applications — tracing, evaluation dashboards, and performance SLOs.- Understanding of caching, batching, and KV-cache strategies for improved throughput and cost optimization.- Exposure to secure tool-use patterns, constrained decoding, and policy-based control mechanisms.Assessment Framework- Portfolio Review: Evaluation of production-ready RAG or agentic systems, including design, architecture, and measurable outcomes.- Technical Exercise: Design an intent router, justify model selection (base vs. fine-tuned), propose chunking/metadata strategies, and define evaluation metrics.- Scenario Discussion: Identify and mitigate failure modes (hallucinations, tool errors, SQL risk).- Governance Evaluation: Demonstrate understanding of access controls, PII handling, audit logging, and red-teaming practices.Qualifications- Bachelor’s or Master’s degree in Computer Science, Artificial Intelligence, Data Engineering, or a related discipline.- 8–12 years of total experience, including at least 3+ years in AI/ML architecture or Gen AI system design.- Proficiency in Python, LLM/RAG frameworks, and modern containerization and CI/CD workflows.- Strong communication, leadership, and problem-solving abilities with a focus on delivering business-impactful AI solutions.