Manpreet S Gurudatta

About Manpreet S Gurudatta

At iMentus Technologies, I serve as a Senior Data Scientist, leading the development of cutting-edge AI and machine learning solutions. My work encompasses designing multi-agent LLM architectures with LangChain and LangGraph frameworks, creating AI-powered resume parsing systems, and engineering conversational database interfaces using transformer models. These innovations support automated workflows, talent acquisition, and natural language-driven database queries. Post Graduate Program in Artificial Intelligence & Machine Learning at The University of Texas at Austin, I specialize in MLOps, NLP, and Computer Vision. Passionate about leveraging technology to solve complex challenges, I focus on creating scalable, production-grade AI systems that integrate seamlessly with diverse enterprise environments.

Experience

1. Senior Data Scientist

   Imentus Technologies Private

   Aug 2023 — Present · Indore, IN

   Multi-Agent LLM Architecture: Built production-ready intelligent agents using LangChain/LangGraph frameworks, integrating Large Language Models with RESTful APIs and micro services for automated workflow orchestration and real-time tool selection•AI-Powered Resume Processing: Developed end-to-end resume parsing pipeline with NLP-based ATS scoring system (80%+ accuracy), implementing ML-driven candidate screening and semantic keyword matching (70%+ precision) for talent acquisition automation•Conversational Database Interface: Engineered natural language to SQL translation system using transformer models, enabling business users to query MySQL databases through conversational AI with secure authentication and query optimization•MCP Server Development: Architected high-performance Model Context Protocol servers using FastMCP framework for MongoDB/MySQL integration, reducing codebase by 80% and improving response latency. Deployed scalable API endpoints for ChatGPT and Claude integration•Cross-Platform Data Integration: Designed distributed database communication layer with event-driven architecture, enabling real-time data synchronization and automated ETL processes across heterogeneous data sources•Production Prompt Engineering: Implemented advanced prompt optimization strategies including few-shot learning, chain-of-thought reasoning, and context window management, achieving 95%+ accuracy in production LLM applications

Education

• The University of Texas at Austin

  Post Graduate Program in Artificial Intelligence & Machine Learning, Data Science