Gauri Thube

About Gauri Thube

Hello! I\'m GauriAnalytically driven professional with 2+ years of extensive experience in Python, LLMs, Generative AI, Machine Learning, Agentic AI and NLP excelling at optimizing AI workflows and data pipelines. Specializes in integrating language models with existing systems, implementing RAG architectures, working with vector databases, AWS cloud architecture and fine-tuning models for specific domains. Combines technical expertise with problem-solving skills to deliver scalable AI solutions.As a quick learner, I am excited about the opportunities that lie ahead and look forward to connecting with like-minded professionals who share my passion for innovation and progress.

Experience

1. Data Scientist

   Deloitte

   Sep 2025 — Present · Mumbai, IN

   Implemented and delivered a GenAI- driven compliance automation system for client Eli Lilly, reducing pharmaceutical clinical trial agreement review time by 60% while maintaining high legal accuracy.• Developed a RAG pipeline using Langchain and Amazon OpenSearch achieving ~85-90% semantic retrieval precision across NGD (Negotiable Guidance Policy Documents) and PNT (Previously Negotiated Terms).• Operationalised the solution on AWS to handle large volumes of NGD documents across multiple countries, enabling scalable review of 100s of multi-jurisdictional contracts per cycle, improving audit traceability by ~80% and reducing manual legal effort by ~50%.• Designed and implemented an Agentic AI powered claims processing system for a Mauritius-based financial services and insurance leader, using LangChain and LangGraph to orchestrate dynamic, stateful and multi-agent workflows achieving ~80% automation in dental claims processing.• Built and deployed LLM-driven specialised agents for document extraction, claim screening, policy eligibility assessment, and real-time customer notifications, reducing manual claim handling effort by ~70%.• Integrated LLMs with enterprise business logic, combining policy rules, historical claims data, and customer profiles to enable explainable decision-making, reducing claim payout errors by ~30%.• Optimized claims workflow, contributing to a 40% reduction in operational costs through intelligent automation and standardisation.• Enabled structured data capture and traceable AI decisions through human-in-the-loop validation, improving compliance adherence and reporting accuracy by ~85%.• Delivered a scalable, modular GenAI architecture, supporting enterprise-wide rollout with ~99% system availability and governed AI adoption.

Education

• SIES Graduate School Of Technology

  Bachelor of Engineering - BE, Information Technology

  2019 — 2023

• St. Xavier's High School

  SSC