Open to opportunities

nirmal murugan

@nirmalmurugan

Message

Data Scientist building RAG-powered GenAI and ML/Computer Vision solutions.

India

Message

What I'm looking for

I’m looking to build and deploy AI products end-to-end—GenAI with RAG, ML/Deep Learning, and scalable APIs—while turning data into actionable insights. I want teams that value clean engineering, evaluation, and practical production impact.

I’m a Data Scientist with 4.2 years of experience building data-driven solutions using Machine Learning, Deep Learning, and Generative AI. I design RAG architecture with Prompt Engineering, Embeddings, and LangChain—using OpenAI, HuggingFace, and LLaMA—backed by vector databases like FAISS and Pinecone for accurate semantic retrieval and contextual responses.

I also build predictive ML models (XGBoost, Random Forest, Logistic Regression) and deep learning pipelines (RNN, LSTM, CNN), including Computer Vision with OpenCV and detectors like YOLOv3, SSD, and R-CNN. From production-ready REST APIs (FastAPI/Flask) to containerization (Docker) and deployment on AWS (S3, EC2, SageMaker), I focus on scalable delivery—like my Personalized Healthcare Chatbot, LLM-powered Fraud Detection assistant, and an Intelligent Loan Approval model achieving ~85–90% accuracy.

Experience

Work history, roles, and key accomplishments

Current

Data Scientist

Current

Microline Informatioion Systems Pvt Ltd

Jan 2022 - Present (4 years 4 months)

Designed and implemented Generative AI solutions using RAG architecture, embeddings, and LLMs with OpenAI, Hugging Face, LLaMA, and LangChain, including vector-database-backed semantic search with FAISS/Pinecone. Built and deployed ML/deep learning models for fraud detection and loan approval, achieving ~85–90% accuracy for loan prediction and reducing manual reporting effort by 40%.