Physics‑Informed Neural Network for Single Particle Model SOC and SOH Prediction
A Physics‑Informed Neural Network (PINN) is a neural network trained not only on data but also on physical laws, typically expressed as differential equations.
A Physics‑Informed Neural Network (PINN) is a neural network trained not only on data but also on physical laws, typically expressed as differential equations.
A Retrieval‑Augmented Generation (RAG) system that ingests documents, creates embeddings, stores them in a vector database, and retrieves relevant context to generate accurate, grounded AI responses using LLMs for search, chat, and knowledge automation.
A fully automated RAG AI agent that ingests new Google Drive files, generates embeddings, stores them in Pinecone, and uses an AI Agent with memory and retrieval to answer chat queries using OpenAI models for accurate, context‑aware responses.
Build an agentic RAG application from scratch by collaborating with Claude Code.
An AI‑powered multi‑action assistant built in n8n that sends emails, fetches real‑time news, and creates meeting invitations from natural language commands. The AI Agent interprets each message, selects the right tool, and executes tasks automatically through Gmail, News API, and Google Calendar.
A workflow that connects WhatsApp with an AI agent using n8n. Incoming messages are processed by an LLM, which generates intelligent responses and sends them back automatically through WhatsApp.
An n8n workflow that reads top AI news from RSS feeds, uses an AI agent to extract and transform the raw articles into clear, readable summaries, and sends the final insights directly to WhatsApp. This creates an automated, real‑time AI news delivery system. 
An automated n8n workflow that fetches top HackerNews stories, extracts raw data, and uses an AI agent to generate clean, readable summaries. The processed insights are sent directly to WhatsApp via the WhatsApp Cloud API, creating a seamless, automated tech‑news delivery system.
Published in 2021 6th International Conference for Convergence in Technology (I2CT), 2021
IEEE conference paper
Recommended citation: Mohammed Nowshad Ruhani Chowdhury, (2021). "Heart Disease Prognosis Using Machine Learning Classification Techniques." Conference 2021 6th International Conference for Convergence in Technology (I2CT). https://ieeexplore.ieee.org/abstract/document/9418181
Published in Journal: IEEE Transactions on Big Data, 2022
Journal paper
Recommended citation: Mohammed Nowshad Ruhani Chowdhury (2022). "ANOVA-based automatic attribute selection and a predictive model for heart disease prognosis." Journal IEEE Transactions on Big Data. 1(2). https://arxiv.org/pdf/2208.00296
Published in Metropolia University of Applied Sciences Journal, 2025
Master’s thesis paper
Recommended citation: Mohammed Nowshad Ruhani Chowdhury. (2025). "Using Open Source LLM Model for Medical Transcription." Metropolia University of Applied Sciences Journal. 1(3). https://www.theseus.fi/bitstream/handle/10024/890628/Chowdhury_Mohammed_Nowshad_Ruhani.pdf?sequence=2
Published in arxiv, 2026
Arxiv Journal
Recommended citation: Mohammed Nowshad Ruhani Chowdhury (2026). "Evaluating Fine-Tuned LLM Model For Medical Transcription With Small Low-Resource Languages Validated Dataset." Arxiv Journal. 1(3). https://arxiv.org/pdf/2603.24772