Pinned Repositories
Agentic_RAG
Agentic RAG System – A multi-agent Retrieval-Augmented Generation (RAG) system built with CrewAI for business intelligence and document analysis. It integrates ChromaDB for document storage and retrieval, real-time web search, and specialized agents for code execution and visualization, enabling automated trend analysis and insights generation
APS-Assignments
BERT-based-Sentiment-Analyzer
Transformer-based sentiment analysis leveraging a BERT architecture for text classification, implemented with PyTorch
Code-Mix-Generation
Developed a deep learning sequence‑to‑sequence model with an attention mechanism, inspired by neural machine translation, to translate monolingual sequences to code‑mixed sequences and vice versa
Colour-blindness-Detection-AFrame-1
SSD13 Project
Conversational-QA-RAG-Chatbot-using-Langchain
Conversational Retrieval-Augmented Generation (RAG) system built with LangChain and Streamlit, enabling PDF uploads and natural chat with document content. Supports chat history for contextual conversations, Groq API for fast inference, and HuggingFace embeddings for document understanding
Data-Analytics-Projects
Data Analytics projects implementing classification (KNN, Decision Trees), clustering (K-Means, Hierarchical, DBSCAN), and frequent itemset mining (Apriori, FP-Growth), along with analysis and insights from large datasets
English-Hindi-NMT-Seq2Seq-Attention
English-Hindi neural machine translation using sequence-to-sequence models with attention mechanism, implemented in PyTorch
Snowflake-Data-Analyst
An AI-powered agentic data analysis system that converts natural language queries into optimized SQL for Snowflake, generates insights, and creates interactive visualizations. Built with LangGraph for agent orchestration, FastAPI for scalable backend services, and Streamlit for an intuitive analytics interface.
stock_market_research_analyst
An agentic AI system leveraging CrewAI for holistic stock market research on Indian-listed companies. Combines fundamental, technical, and market news analysis using specialized AI agents and custom data tools to deliver actionable Buy/Sell/Hold recommendations and automated research reports.
vishalpatel72's Repositories
vishalpatel72/Snowflake-Data-Analyst
An AI-powered agentic data analysis system that converts natural language queries into optimized SQL for Snowflake, generates insights, and creates interactive visualizations. Built with LangGraph for agent orchestration, FastAPI for scalable backend services, and Streamlit for an intuitive analytics interface.
vishalpatel72/Agentic_RAG
Agentic RAG System – A multi-agent Retrieval-Augmented Generation (RAG) system built with CrewAI for business intelligence and document analysis. It integrates ChromaDB for document storage and retrieval, real-time web search, and specialized agents for code execution and visualization, enabling automated trend analysis and insights generation
vishalpatel72/APS-Assignments
vishalpatel72/BERT-based-Sentiment-Analyzer
Transformer-based sentiment analysis leveraging a BERT architecture for text classification, implemented with PyTorch
vishalpatel72/Code-Mix-Generation
Developed a deep learning sequence‑to‑sequence model with an attention mechanism, inspired by neural machine translation, to translate monolingual sequences to code‑mixed sequences and vice versa
vishalpatel72/Colour-blindness-Detection-AFrame-1
SSD13 Project
vishalpatel72/Conversational-QA-RAG-Chatbot-using-Langchain
Conversational Retrieval-Augmented Generation (RAG) system built with LangChain and Streamlit, enabling PDF uploads and natural chat with document content. Supports chat history for contextual conversations, Groq API for fast inference, and HuggingFace embeddings for document understanding
vishalpatel72/Data-Analytics-Projects
Data Analytics projects implementing classification (KNN, Decision Trees), clustering (K-Means, Hierarchical, DBSCAN), and frequent itemset mining (Apriori, FP-Growth), along with analysis and insights from large datasets
vishalpatel72/English-Hindi-NMT-Seq2Seq-Attention
English-Hindi neural machine translation using sequence-to-sequence models with attention mechanism, implemented in PyTorch
vishalpatel72/IIIT-Assignment
All the assignments solved as a student of IIIT
vishalpatel72/NLP-Assignments
Implemented and compared Statistical and Neural Language Models. Built N‑gram models with Kneser‑Ney & Witten‑Bell smoothing and a Neural Language Model, evaluated using sentence‑level and corpus‑level perplexity scores on multiple text corpora to analyze performance differences between statistical and neural approaches
vishalpatel72/Prescription
vishalpatel72/SMAI-Assignments
Assignments covering machine learning, deep learning, and representation learning: regression, models done as a part of my MTech at IIIT-H
vishalpatel72/SSD-Assignments
vishalpatel72/stock_market_research_analyst
An agentic AI system leveraging CrewAI for holistic stock market research on Indian-listed companies. Combines fundamental, technical, and market news analysis using specialized AI agents and custom data tools to deliver actionable Buy/Sell/Hold recommendations and automated research reports.
vishalpatel72/vishalpatel72.github.io
vishalpatel72/Doctor-Diagnosis-AI-Agent
Doctor Diagnosis AI Agent – A Streamlit-based web application for medical image analysis using OpenAI’s GPT-4 with vision. Supports AI-powered image diagnostics, explainable AI, collaborative discussion, report Q&A, and PDF statistical report generation to assist healthcare professionals in diagnosis
vishalpatel72/end-to-end-text-summarizer
End-to-end text summarization pipeline demonstrating fine-tuning of Transformer-based models on custom data, integrating with a FastAPI REST API, and deploying to AWS using Docker and GitHub Actions CI/CD. Includes a modular codebase for training, evaluation, and production deployment.
vishalpatel72/Mini-SQL-Engine
A lightweight SQL engine implemented in Python that parses and executes a subset of SQL queries via the command line. It supports core SQL features like SELECT, WHERE, GROUP BY, ORDER BY, DISTINCT, and simple joins on CSV-based datasets, providing an educational insight into how SQL engines work internally.
vishalpatel72/Wikipedia-Search-Engine
A Wikipedia search engine using Block Sort-Based Indexing (BSBI) with field-specific query support and TF-IDF ranking, built during my MTech at IIIT-Hyderabad.