Data Scientist

Data Scientist – GenAI Expert AI-Powered Product Engineering Team Role Overview We're looking for a Data Scientist with deep GenAI expertise to help design and optimize LLM-powered pipelines for AI-driven products. This role sits at the intersection of prompt engineering and production Gen AI and ML systems. Key Responsibilities Design, build, and optimize retrieval-augmented generation (RAG) architectures and LLM orchestration pipelines Own prompt engineering and fine-tuning efforts to improve output quality, tone, and consistency Benchmark and evaluate multiple LLM providers for quality, cost, and fit for purpose Build evaluation frameworks for content quality, factual accuracy, and compliance adherence where relevant Diagnose and resolve issues in LangGraph based AI pipelines, including prompt structure, context retrieval, and output consistency Work with engineering to productionize models and pipelines, including vector databases and retrieval systems Partner with stakeholders to translate business and quality requirements into technical pipeline logic Support workflow orchestration with AI-driven quality checks and human-in-the-loop review steps Required Skills & Experience 6-10 years of experience in data science/applied ML, with at least 1-2 years hands-on with LLMs/GenAI in production Strong Python skills; experience with LangChain, LangGraph, or equivalent orchestration frameworks Hands-on experience with RAG architectures, vector databases (e.g., PgVector, Weaviate, FAISS etc), and embedding models Experience with major LLM APIs (OpenAI/Azure OpenAI, Anthropic, Google Vertex AI, or other providers) Strong prompt engineering skills, including structured prompting and few-shot design Solid understanding of NLP evaluation metrics and methods for assessing generative output quality Good to Have Experience in regulated or compliance-sensitive domains Multilingual content experience or fluency-tuning for non-English outputs Experience with content or product platforms integrated into broader marketing/tech stacks Familiarity with model comparison/benchmarking methodologies across multiple LLM providers