Successfully developed a real-time AI backend using FastAPI and WebSockets with low-latency LLM streaming. Implements conditional tool calling, multi-turn state management, and Supabase-based persi…

This repository contains LangGraph tutorial code for building complex LLM-powered applications. It showcases practical examples of using LangGraph to manage multi-step workflows and stateful logic …

Successfully developed a Travel Planner Assistant powered by LangGraph and MCP, integrating real-time flights, hotels, weather, places, and timezone tools. Features modular MCP servers, OpenAI agen…

Successfully developed a real-time AI-powered Insurance Advisor CoPilot that converts customer queries into actionable insights, guiding users effortlessly through insurance coverage. It delivers i…

Successfully established a robust Pistol Object Detection system using Faster R-CNN with ResNet50 RPN, capable of accurately identifying and localizing firearms in images. Integrated the model into…

Successfully developed a CNN-based image classification model to accurately identify and categorize various animal species.

Successfully developed a CNN-based model to predict age groups such as young, middle and old from facial images, enabling accurate age classification for applications in security, personalization, …

Successfully developed a Multi-Source Knowledge Intelligence Agent - an advanced agentic AI application that integrates multi-format document and web data processing with LLM-powered analytics and …

An end-to-end AI Financial Research & Advisory Agent built using LangGraph, LLMs, and machine learning for automated market analysis, compliance validation, and investment advisory. Integrates real…

This repository contains an implementation of the Q-learning algorithm using Gymnasium environments. It demonstrates training, evaluation, and visualization of an agent’s policy in a simple reinfor…

This repository covers extensive tutorials on how to integrate LangSmith with LangChain with LangGraph to incorporate observability, monitoring, alerting, evaluation, etc. within complex LLM workfl…

This project encompasses end-to-end ETL and ML pipeline development. Data ingestion from TMDB API covered top-rated, current, upcoming, and popular movies with genres. Performed EDA to derive sever…

Successfully performed a comprehensive analysis of trader behavior in relation to Bitcoin market sentiment (Fear vs. Greed) using real-world trading and sentiment datasets. Key data-driven insights…

Successfully built a crop yield prediction system using advanced machine learning in R, which leverages agro-environmental, soil, and weather data to forecast agricultural yield (in tons per hectar…

Successfully developed a robust customer churn prediction system using advanced machine learning models in R, leveraging techniques like logistic regression, decision trees, random forests, boostin…

Successfully developed a real-time soccer player re-identification and tracking system using YOLOv11 and Deep SORT with TorchreID-based embeddings.

This is my portfolio website.

Successfully developed an object detection model using YOLOv8 to detect the images of fruits.

Successfully developed an instance segmentation model using Mask R-CNN to detect and segment brain tumors from MRI scans with pixel-level precision.

Developed an instance segmentation model using Mask R-CNN to accurately identify and segment germinated seeds in high-resolution seed images.

Successfully developed a Mask R-CNN-based instance segmentation model to detect and segment various types of road lane markings from complex street images.

Successfully developed an object detection model using YOLOv8 to identify safety equipment and personnel on construction sites, enhancing on-site safety compliance and monitoring.

Successfully developed an object detection model using Faster R-CNN to detect safety helmets and ensure compliance at construction sites by accurately localizing helmets and personnel in real-time …

Successfully developed an object detection model using Faster R-CNN to detect and localize wind turbines in aerial imagery, aiding in automated monitoring and infrastructure assessment.

Successfully developed an object detection model using Faster R-CNN to detect and classify traffic signs in road images, enhancing autonomous driving and intelligent transportation systems.