Portfolio
Practical applications of research in AI/ML and intelligent systems.
A Python-based Retrieval-Augmented Generation (RAG) system for document question-answering. Built with FastAPI, HuggingFace transformers, and FAISS vector search. Supports PDF, TXT, DOCX, MD ingestion with both REST API and Streamlit web interface.
A deep reinforcement learning approach to optimize smart home energy usage while preventing cartel-like behavior in peer-to-peer energy markets. Implements DDPG to optimize HVAC operation, battery management, and price-setting strategies with anti-cartel mechanisms for market fairness.
A state-of-the-art AI text detection system built by fine-tuning Google's Gemma3-4B model using Unsloth optimizations and the RAID dataset. Detects AI-generated text across 11 LLMs and multiple domains with memory-efficient training on consumer hardware.
Production-ready ML pipeline for digit recognition using Kubeflow Pipelines, MLflow, and KServe. Demonstrates complete MLOps workflow from data acquisition to model serving with experiment tracking, containerized components, and infrastructure as code.
A lightweight neural network pruning library built on Microsoft NNI for PyTorch models. Implements L1 norm pruning to reduce model size and computational requirements for edge deployment while maintaining performance.
Comprehensive comparative study of decentralized vs centralized federated learning architectures. Extends the Fedra P2P framework with Flower comparison, hierarchical FL implementation, and experiments across synthetic and real-world datasets including LSTM-based traffic prediction.
A Deep Q-Network (DQN) implementation for training an AI agent to play the classic Snake game using Reinforcement Learning. Uses PyTorch for the neural network, Gym for the environment, and Pygame for rendering with comprehensive training visualization.
Developed a distributed energy management simulation framework modeling multiple household HVAC systems and storage units. The system employs reinforcement learning algorithms to optimize energy distribution across the network while maintaining temperature constraints.
Let's discuss research or development opportunities.
