Available for AI/ML opportunities
AI/ML Engineer + Applied Researcher
Hi, this is Refat Ahmed, an EEE graduate specializing in applied AI/ML, with experience in building real-world machine learning and automation systems.
I am currently pursuing an M.Sc. in Computer Science & Engineering at the Military Institute of Science and Technology (MIST), while building applied AI/ML systems across NLP, computer vision, healthcare prediction, and automation.
Military Institute of Science and Technology (MIST), Dhaka.
Islamic University of Technology (IUT), Gazipur, Dhaka. CGPA: 3.38.
Notre Dame College, Dhaka. GPA: 5.00.
Professional Work
Applied engineering experience across NLP systems, AI agents, object detection, and academic healthcare research.
Factoryze
United International University
Toolbox
Tools and frameworks used across production prototypes, academic experiments, and research workflows.
PyTorch, TensorFlow, Scikit-learn, TorchVision, YOLO, OpenCV, SHAP
LangChain, LangGraph, Ollama, Hugging Face Transformers, Groq LLM
Python, FastAPI, RESTful APIs, SQL, Git, GitHub
AWS SageMaker, N8N, MATLAB, Simulink
Showcase
A portfolio of AI agents, NLP systems, computer vision models, and machine learning experiments. For workflow automation work, visit the dedicated n8n page.
Real-time hand-raise detection system for corporate office, built with a custom YOLOv11 object detection pipeline and video inference.
RESTful chatbot API that produces natural, context-aware responses to product queries using a Groq-hosted large language model and external product data.
Local AI assistant that interprets natural language event requests and maps them into Google Calendar actions with a two-stage Ollama and LangGraph pipeline.
Cloud-ready calendar automation system for conversational event creation, cancellation, and rescheduling from natural language instructions.
Knowledge distillation workflow for Banglish imperative sentence classification, focused on faster inference in low-resource environments.
Hybrid NLP classifier for identifying deadline-related expressions from Banglish text, supporting task extraction and office automation workflows.
Chatbot exploration using open-source Hugging Face models for conversational response generation and LLM experimentation.
Vision project for detecting sign-language gestures with image processing and machine-learning techniques.
Comparative deep-learning study using CNN, DNN, and LSTM models on graphical ECG signal representations for arrhythmia detection.
Computer vision project that extracts, aligns, and blends video frames into panoramic images for footage analysis and wide-view generation.
Machine-learning project for predicting customer retention risk and identifying business signals that can support proactive fintech customer engagement.
Exploratory analysis project focused on extracting business and product insights from fintech data using notebook-driven analytics.
Research Experience
Research focus in biomedical engineering and applied machine learning, with emphasis on developing and translating models for industrial and healthcare applications
Thesis
Thesis focused on comparative evaluation of models, feature analysis, and implementation of practical stroke risk prediction workflows with a Streamlit-based interface, using real-world clinical data collected from Faridpur.
Worked on the “iCRP: An Intelligent CVD Risk Profiling for early identification of MI and Stroke in Bangladesh” project, contributing to exploratory data analysis, initial literature review drafting.
Academic Output
Research outputs include 1 Q1 journal publication, 1 international conference presentation in India, and 2 conference abstracts
Zaman, M., Hridhee, R.A., Bhuiyan, R.A. et al. BMC Public Health 25, 1833 (2025).
DOIRefat Ahmed Bhuiyan, Farhana Sarker, Margub Aref Jahangir, et al.
DOI
Bhuiyan, R.A., Hamid, S.A., Hassan, M.I., et al.
2025 Global Public Health Systems Conference (GPHS), Colombo, Sri Lanka
Bhuiyan, R.A., Hamid, S.A., Hassan, M.I., et al.
Addis Ababa, Ethiopia, Oct 2025
Recognition
Awards, competitions, and conference milestones from applied AI, healthcare innovation, and research presentation work.
Winner
Developed and presented a prototype for AI-based early detection of endometriosis.
Finalist
Advanced through EDA, statistical analysis, predictive modeling, and deep learning stages.
Conference Presentation
Presented research findings and technical contributions at an academic conference in Kerala, India.
Workflow Automation
n8n automation systems, workflow screenshots, trigger logic, connected apps, and production-ready automation case studies.
Automation Portfolio
Explore the n8n portfolio featuring workflow visuals, node architecture, integrations, automation outcomes, project documentation, and client implementations.