About the role
Giniminds seeks a skilled LLM Data Scientist to design, develop, and optimize agentic AI systems and enterprise AI platforms. You will implement LangChain, LangGraph, and LLM flows, build multi-agent workflows, and collaborate with product, engineering, and AIOps teams to deploy production-grade AI capabilities.
The role requires 2+ years in data science and NLP, strong Python and cloud experience, and a hands-on approach to vector databases, RAG design, and model tuning.
Build, evaluate, and optimize LLMs, knowledge-grounded models, and agentic pipelines. Develop multi-agent systems, tool-based agents, retrieval-augmented agents, and workflow automations. Implement LangChain, LangGraph, LLM flows or equivalent frameworks for orchestration. Fine-tune open-source LLMs (Llama, Mistral, Gemma) for domain-specific tasks. Design and deploy end-to-end AI pipelines from ingestion to inference & monitoring. Integrate LLM-based systems with external APIs, SaaS tools, and enterprise systems. Build reusable modules, prompt libraries, agent frameworks, and evaluation suites. Work closely with MLOps teams for optimization, caching, latency reduction, and cost control. Perform advanced prompt engineering, RAG tuning, embeddings design, and model evaluation. Design experiments, analyze model performance, and implement continuous improvement cycles. Develop structured datasets for LLM training, synthetic data generation, and data quality validation. Partner with architects to design scalable AI platform architecture. Translate business requirements into technical designs and AI workflows. Document solutions, maintain repos, and follow reproducibility & governance. 2+ years of hands-on Data Science and NLP experience. Proficient in Python, LangChain, LLM orchestration, and API integration. Experience with vector databases and retrieval-augmented architectures. Knowledge of model evaluation, embeddings, and RAG. Familiar with cloud platforms and observability tooling. Experience deploying AI pipelines and tooling in production. Strong collaboration with product and engineering teams. Understanding of governance and reproducibility in AI systems. Ability to design scalable AI workflows and agent-based automation.