AI/ML Engineer · Machine Learning · Deep Learning · Generative AI · MLOps

Building end-to-end AI systems that move from data → models → APIs → applications

I'm an AI/ML Engineer focused on developing practical machine learning, deep learning, and Generative AI applications.

My work spans the AI development lifecycle, including data preprocessing, model training, evaluation, experiment tracking, REST APIs, and interactive applications.

• 🔭 Building end-to-end ML, DL, and GenAI projects
• 🤖 Exploring Machine Learning, Computer Vision, NLP, and Retrieval-Augmented Generation
• ⚙️ Developing API-driven AI applications with FastAPI and Streamlit
• 📊 Working with model experimentation and tracking using MLflow
• 🌱 Expanding my knowledge of advanced RAG, Hugging Face, Agentic AI, and LLMOps
• 🎯 Open to AI/ML, Machine Learning, Applied AI, and Generative AI internship opportunities

Deep learning computer vision project for industrial defect classification using a custom ResNet-50 architecture.

🔗 View Repository

Key Features

• Custom ResNet-50 implementation using TensorFlow/Keras
• Industrial image classification using the MVTec AD dataset
• Image preprocessing and data augmentation
• Model evaluation and performance analysis
• Experiment tracking with MLflow
• FastAPI inference backend and Streamlit interface
• MLOps-oriented project structure

Tech Stack: Python · TensorFlow · Keras · Computer Vision · MLflow · FastAPI · Streamlit

A Retrieval-Augmented Generation application that answers questions using uploaded medical PDF documents.

🔗 View Repository

Key Features

• Upload medical PDFs and build a searchable knowledge base
• Semantic retrieval using FAISS
• Context-grounded question answering
• Document-based responses to support source-grounded information retrieval
• Interactive Streamlit interface
• API key management through environment variables

Tech Stack: Python · RAG · LLMs · LangChain · FAISS · Embeddings · Streamlit · PDF Processing

Note: This is an experimental document-question-answering application, not a substitute for professional medical advice.

An end-to-end machine learning application for predicting customer responses to bank marketing campaigns.

🔗 View Repository

Key Features

• End-to-end preprocessing and model training pipeline
• Multiple classification algorithms and hyperparameter tuning
• Class imbalance handling using SMOTE
• Experiment tracking with MLflow
• Model persistence using Joblib
• REST API with FastAPI
• Interactive prediction interface with Streamlit

Tech Stack: Python · Pandas · NumPy · Scikit-learn · XGBoost · SMOTE · MLflow · FastAPI · Streamlit

Areas: RAG · Embeddings · Vector Search · LLM Applications · Prompt Engineering · NLP · Agentic AI

Focus Areas: ML Pipelines · Experiment Tracking · Model Evaluation · REST APIs · Model Deployment · Reproducibility

• Advanced and hybrid RAG architectures
• Multimodal RAG
• LLM fine-tuning with Hugging Face
• Agentic AI and multi-agent systems
• MLOps and LLMOps
• CI/CD for machine learning applications

• AI/ML Engineering
• Machine Learning and Deep Learning
• Generative AI and RAG
• Computer Vision and NLP
• MLOps and LLMOps

I'm interested in collaborating on machine learning, deep learning, Generative AI, RAG, computer vision, and production-oriented AI projects.

Building AI systems one project at a time 🚀