
Data Analyst with 1+ year of experience in building RAG systems, AI chatbots, NLP-to-SQL applications, and Power BI dashboards. Skilled in LLMs, semantic search, vector databases, machine learning, and analytics-driven enterprise solutions.
Regulatory Document Intelligence Platform | RAG-Based AI System
• Built an end-to-end RAG-based regulatory document intelligence platform for citation-backed question answering over large PDF corpora..
• Developed a hybrid PDF extraction pipeline using PyMuPDF and Vision LLMs for accurate text and table extraction with page-level traceability.
• Implemented embedding, chunking, and ChromaDB(Vector Database) ingestion pipelines with enriched metadata for efficient semantic retrieval.
• Engineered a hybrid retrieval system combining vector search, BM25, and RRF reranking to improve retrieval accuracy for regulatory and legal documents.
Procurement Analytics & AI Chatbot Platform | Power BI & GenAI
• Developed a Procurement Analytics Dashboard Suite using Power BI, covering spend analysis, vendor performance, purchase orders, and inventory insights for client reporting and decision support.
• Designed interactive dashboards with KPI tracking, trend analysis, and drill-down capabilities to improve business visibility and facilitate data-driven decision making.
• Built AI-powered chatbot to convert natural language queries into SQL, retrieve procurement data, and generate LLM-based insight summaries for streamlined data access.
• Built an AI-powered chatbot that converts natural language queries into SQL, retrieves procurement data, and generates LLM-based insight summaries.
Data visualization
Power BI
Sql
Python
Machine Learning
Natural Language Processing
Generative AI
Retrieval Augmented Generation
Agentic AI
Langchain & LangGraph
Retrieval Augment Generation
API Creation
LLM Integration
AI chatbot development
Predictive Maintenance System | Machine Learning & MLOps
• Developed a predictive maintenance solution to classify machine failures using sensor and operational data.
• Performed feature engineering, RFE-based feature selection, and hyperparameter optimization using Scikit-learn and Random Forest.
• Built end-to-end MLOps pipelines with MLflow, FastAPI, and Docker for experiment tracking, deployment, and model serving.
• Deployed the solution on Google Cloud Platform (GCP) and implemented data drift monitoring for model performance tracking.
• Achieved 99.9% accuracy and 0.984 ROC-AUC, enabling proactive maintenance and reduced operational downtime.
Reading