Lavanya Mupparaju

I am an AI/ML Engineer with 3 years of experience in designing and delivering data-driven solutions across banking, insurance, E-commerce, healthcare, and legal domains. I specialize in Generative AI, Agentic AI, Machine Learning, and NLP, with hands-on expertise in advanced techniques such as Retrieval-Augmented Generation (RAG), finetuning, quantization, prompt engineering, agent development, and Model Context Protocol (MCP) development. I have a proven track record of building predictive models, developing scalable AI systems, and deploying intelligent solutions that enhance decision-making, optimize operations and deliver measurable business impact.

1. Intelligent End-to-End Travel Planning and Decision Support System

• Technologies: Python, Streamlit,FastAPI,Langgraph,Multi agent system,langchain,OpenAI GPT models,RAG,chromadb,FAISS,MCP

•Developed an end-to-end AI-powered travel planning platform using Python, FastAPI, Streamlit, LangChain, and Large Language Models (LLMs).

• Designed a Multi-Agent Architecture consisting of specialized agents for itinerary generation, flight recommendations, hotel suggestions, weather analysis, destination discovery, and budget estimation.
• Implemented a Supervisor Agent to intelligently route user requests and coordinate communication among multiple AI agents.
• Integrated Real-Time Web Search capabilities to retrieve up-to-date travel information, local attractions, transportation details, events, and destination insights.
• Utilized Model Context Protocol (MCP) to enable seamless interaction between AI agents and external tools, APIs, and data sources.
• Built Retrieval-Augmented Generation (RAG) pipelines using vector databases to enhance response accuracy and provide context-aware travel recommendations.
• Developed dynamic day-wise itinerary generation workflows based on user preferences, travel duration, budget constraints, and destination-specific information.
• Implemented conversational AI capabilities allowing users to interact naturally and receive personalized travel plans through a chat-based interface.
• Designed scalable backend APIs using FastAPI and integrated them with a responsive Streamlit frontend for real-time user interactions.
• Optimized agent orchestration, prompt engineering, and tool-calling workflows to deliver comprehensive travel plans with personalized recommendations and cost estimates.

2. Automated Invoice Extraction System

• Technologies: Python, Llama-Index, LangChain, GPT-4o, Azure Form Recognizer, Azure Cognitive Search, Prompt, Fast API, Angular

• Developed an AI-driven solution using GPT-4o and Llama-Index to automate entity extraction from invoices across varying templates and geographies.

• Automated data categorization and structured extracted data into relational tables, ensuring high accuracy and scalability.

• Built a logging mechanism to capture ambiguities and flag exceptions for manual review, enhancing data reliability.

• Conducted extensive testing and implemented backup/recovery strategies to maintain system reliability under high invoice volumes.

3. Enterprise Knowledge Accelerator

• Technologies: Python, FastAPI, OpenAI GPT-4o, PyPDF2, REST APIs, Uvicorn, Prompt Engineering, Agentic AI

• Developed an Agentic AI-based PDF Learning Platform using Python and FastAPI that enables users to upload PDF documents and automatically generate personalized learning resources.

• Implemented a PDF processing pipeline to validate uploads, extract document text, and prepare content for AI- driven analysis.

• Designed an AI Agent workflow that analyzes document content, identifies key topics and complexity levels, and creates an execution plan for downstream processing.

• Built a tool orchestration framework where the agent dynamically decides and invokes specialized tools such as Summary Tool, Notes Tool, Quiz Tool, and Interview Questions Tool based on document analysis.

• Integrated OpenAI GPT models within individual tools to generate contextual summaries, study notes, quizzes, and interview preparation materials from extracted PDF content.

• Architected the solution using a modular FastAPI design pattern, separating routers, services, and tools to improve maintainability, scalability.

4. TalentMatch – Intelligent Resume Screening and Skill Gap Analysis Platform

• Technologies: Python, Streamlit, NLP, Sentence Transformers, Scikit-Learn, PyPDF2, Cosine Similarity, Pandas

• Developed an NLP-based resume screening platform that compares candidate resumes with job descriptions and evaluates candidate suitability using semantic similarity analysis.

• Built a PDF resume processing pipeline that extracts and cleans resume content, enabling accurate analysis of candidate profiles from uploaded documents.

• Implemented sentence embedding techniques using transformer-based language models to convert resumes and job descriptions into vector representations for contextual comparison.

• Utilized cosine similarity algorithms to calculate candidate-job matching scores, helping recruiters quickly identify suitable applicants and reduce manual screening efforts.

• Designed a skill gap analysis module that identifies matched skills between the resume and job description while highlighting missing skills required for the target role.

Developed an interactive Streamlit-based user interface that allows users to upload resumes, enter job descriptions, view matching percentages, and receive personalized skill improvement recommendations.