Job Title

Key Responsibilities:- Lead the end-to-end design and development of Generative AI applications using LLMs. - Architect intelligent systems using LangChain / LangGraph for multi-agent workflows, planning, task execution, verification, and orchestration. - Implement RAG pipelines using embeddings, hybrid search, semantic indexing, and reranking. - Design and optimize Vector DB structures (FAISS, Qdrant, Pinecone, Weaviate, Chroma) for high-performance retrieval. - Create knowledge ingestion pipelines including chunking, preprocessing, OCR and document embedding strategies. - Build scalable backend services using Python / FastAPI / Node integrated with LLMs (OpenAI, Anthropic, Gemini, Llama). - Own POCs, architecture decisions, prompt engineering frameworks, and performance tuning. - Drive experimentation of LLM models, fine-tuning, LoRA, embedding selection and evaluation metrics. - Collaborate with cloud and product teams to deploy scalable GenAI solutions on AWS / GCP / Azure. - Mentor and guide engineers, review code, and set best practices for model usage & observability. - Stay current with research trends in GenAI, multi-agent systems and knowledge reasoning.Required Skills & Expertise- 5–10+ years overall development experience, 1-2+ years in Generative AI. - Strong expertise in LangChain & LangGraph for orchestration, agents, tools and memory management. - Hands-on experience with Vector Databases (Qdrant / Pinecone / FAISS / Chroma / Weaviate). - Deep understanding of LLM architecture, embeddings, RAG, token optimization & context strategies. - Experience with Multi-Agent Frameworks (Planner, Worker, Verifier, Summarizer). - Strong programming experience in Python (preferred) or NodeJS. - Familiarity with LLM model providers – OpenAI, Anthropic, Gemini, HuggingFace, Meta Llama. - Experience with cloud-native deployment – Docker, Kubernetes, CI/CD. - Understanding of prompt engineering, safety guardrails, hallucination prevention & evaluation. - Ability to lead projects, conduct research, run experiments and deliver production-ready systems.