Join Iris Software, a rapidly expanding IT services company recognized among India's Top Workplaces in the IT. We are seeking a GenAI Architect to spearhead the design and implementation of cutting-edge Generative AI solutions for diverse enterprise needs. This role demands profound expertise in LLMs, RAG, vector databases, and agentic AI frameworks, complemented by experience in cloud-native deployments and CI/CD automation. You will be instrumental in collaborating with stakeholders, data/ML teams, and engineering departments to deliver robust, secure, and production-ready GenAI platforms and applications.

Architect and design comprehensive GenAI solutions, including RAG pipelines, agentic workflows, and multimodal use cases. Translate complex business challenges into actionable GenAI use cases, solution blueprints, and strategic roadmaps. Develop sophisticated retrieval systems utilizing embeddings, chunking strategies, metadata filters, reranking, and rigorous evaluation metrics. Select and integrate vector databases, optimizing indexing, retrieval performance, and relevance. Construct agentic systems using frameworks like LangChain and LangGraph. Define scalable and secure cloud architecture patterns for GenAI deployments. Drive productionization through efficient CI/CD pipelines and containerization best practices. Uphold responsible AI principles, ensuring security, governance, privacy, compliance, and effective monitoring. Provide essential technical leadership, mentorship, and guidance on best practices to engineering teams. Collaborate with product and delivery teams to align solutions with project timelines and desired business outcomes.

Possess deep knowledge of GenAI algorithms and LLM concepts, including prompting techniques, fine-tuning versus RAG, embeddings, context windows, token limitations, and hallucination control. Showcase experience in designing enterprise GenAI use cases such as document Q&A, copilot development, summarization, advanced search functionalities, workflow automation, and knowledge assistant creation. Demonstrate a solid understanding of evaluation techniques, focusing on groundedness, relevance, faithfulness, and cost-performance trade-offs. Have hands-on experience with vector databases and similarity search, encompassing embeddings, indexing, ANN search, hybrid search, and metadata filtering. Proficiency with tools like Pinecone, FAISS, Weaviate, Chroma, Milvus, or Azure AI Search is essential. Strong working knowledge of agentic frameworks, including LangChain, LangGraph, and MCP, is required, alongside expertise in tool/function calling, memory management, planning, multi-agent workflows, and guardrails. Experience with cloud services and architecture tailored for GenAI workloads, covering compute, networking, storage, IAM/security, and logging/monitoring, is crucial. Familiarity with cloud components supporting AI/ML solutions, with a preference for managed services, is expected. Implement CI/CD pipelines for GenAI applications and services. Demonstrate a strong grasp of deployment patterns such as containers (Docker), orchestration (Kubernetes), API deployment, and model endpoint integration. Understand DevOps/MLOps practices, including testing, observability, rollback strategies, scaling, and cost management.