Projects

University LLM Q&A System

Developed an advanced Q&A system using large language models for university applications.

Topic Categorization Engine with LLMs

Developed an NLP-based system using BERT to classify search queries into relevant topics. Built an end-to-end pipeline for data ingestion, model training, and real-time inference.

Sentiment Analysis for Stock Market & E-commerce

Created an AI-driven sentiment analysis model to evaluate stock-related discussions and predict trends based on user emotions. Also applied similar techniques for e-commerce sales forecasting.

Google Trend Prediction Platform

Built a system to process 100M+ monthly data points from Google SERP, using clustering algorithms and machine learning models to forecast keyword trends.

Kafka-Based Real-Time Data Streaming

Designed and implemented real-time data pipelines using Apache Kafka to integrate multiple sources for real-time business insights.

Large-Scale Data Deduplication

Developed a system to detect and remove duplicates from large-scale CRM data, integrating data from multiple sources and storing results in BigQuery.

Web Crawlers for Social & Business Platforms

Built and maintained scalable crawlers for Quora, Reddit, Twitter, and LinkedIn, collecting millions of records with robust error handling.