I build practical AI systems across Generative AI, Retrieval-Augmented Generation, Computer Vision, and Machine Learning.
Currently exploring Agentic AI, MCP, Azure AI, and MLOps to build and deploy production-ready AI applications.
I’m an AI Engineer with an MSc in Artificial Intelligence from the University of Surrey and a B.Tech in Artificial Intelligence & Data Science.
My experience spans Computer Vision, Deep Learning, Natural Language Processing, and Generative AI. I’ve worked on image retrieval and segmentation systems as well as RAG applications using LLMs, embeddings, and vector databases.
I enjoy turning AI concepts into practical applications, from model development and experimentation to APIs, automation, containerization, and cloud deployment.
Python SQL
PyTorch Hugging Face Transformers timm OpenCV scikit-learn
LLMs RAG LangChain OpenAI Sentence Transformers
Pinecone ChromaDB
FastAPI PostgreSQL SQLAlchemy Docker Docker Compose
n8n Azure Git GitHub
Agentic AI MCP Azure AI / Microsoft Foundry MLOps
A deep learning-based image retrieval system developed as part of my MSc dissertation. The project explores deep feature representation and retrieval using DOLG feature fusion, ArcFace loss, and multiple backbone architectures.
Key areas: Image Retrieval • Deep Learning • Feature Fusion • Model Evaluation • Interpretability
Tech: PyTorch DOLG ArcFace ConvNeXt EVA Grad-CAM mAP
A RAG application that combines document processing, semantic retrieval, embeddings, and large language models to provide context-aware responses from a custom knowledge base.
Key areas: RAG • Semantic Search • Embeddings • LLMs • Document Processing
Tech: Python LangChain OpenAI Sentence Transformers Vector Database
A full-stack application combining a React frontend with a FastAPI backend, PostgreSQL database, Docker containerization, and Azure deployment.
Key areas: API Development • Backend Engineering • Containerization • Cloud Deployment
Tech: FastAPI React PostgreSQL Docker Azure
An automated Retrieval-Augmented Generation workflow built with n8n that processes documents from Google Drive, generates embeddings using OpenAI, stores them in Pinecone, and uses an AI Agent to retrieve relevant context for user queries.
Key areas: RAG • AI Agents • Workflow Automation • Semantic Search • Document Processing
Tech: n8n OpenAI Pinecone Google Drive AI Agents