Data Scientist
CurrentData Scientist Intern at Global Technology Legal Compliance Team, SAP Canada.1. Developed an automated AI-powered labeling pipeline for 2000+ non-standard open-source licenses,identifying permissions and obligations.2. Experimented with 20+ prompt versions and 5+ Large Language Models, auto logging evaluation results using MLFlow, achieving 70% accuracy and over 0.6 inter-annotator agreement.3. Enhanced the stability and consistency of LLM responses with structured output by replacing JSONmode with function calling and Pydantic.4. Deployed a Retrieval-Augmented Generation (RAG) system using LangChain and vector databases(Weaviate) to feed domain-specific knowledge to LLMs.