Welcome!
đ⤠to my digital self, My little world in digital cyberspace...
I'm Md Faisal Kabir â an Artificial Intelligence, Machine Learning & Deep Learning Researcher & Practitioner! Driven by a passion for building intelligent systems that blend logic with creativity to craft impactful experiences...
"Passion and diligence are two traits that I believe best reflect me as an individual. I believe this is the key to being successful in any endeavour one sets their mind to. I love to generate new ideas and devise feasible solutions to broadly relevant problems. I enjoy embracing the lessons learned from failure, standing up and continue to grow..."
Get in touch faisalkabir.cse@gmail.com
My research focuses on Computer Vision, with a particular emphasis on Explainable AI (XAI) and Large Language Models (LLMs). One of my greatest strengths is my ability to translate conceptual ideas into practical, working computer programs.
Currently I am working as a Software Engineer (Machine Learning) at Genuity Systems Limited and Research Scientist (Computer Vision & AI Initiatives) at Kharagpur Learning, Imaging & Visualization (KLIV) research group â IITKgp India, developing methods for "Identifying and Segmenting Optical Images" to enhance automated interpretation across different modalities. This aims to improve augmented reality, predictive systems and patient comfort in medical settings.
Being a Artificial Intelligence enthusiast, I enjoy bridging the gap between engineering and AI — combining deep technical expertise with a human-centered mindset to create intelligent systems. My goal is to build scalable, efficient applications that not only perform flawlessly under the hood but also deliver seamless, engaging, and pixel-perfect user experiences.
When I'm not in front of a computer screen, I'm probably traveling somewhere, watching movies, reading books, or crossing off another item on my bucket list. I am always happy to talk about research, technology, outdoor activities, fieldwork, STEM equity or science communication. Whether looking for a research collaborator, do not hesitate to email me and/or follow me on social media.
- Python
- C
- C++
- C#
- HTML
- CSS/Sass
- PHP
- SQL
- Keras
- Tensorflow
- Pytorch
- OpenCV
- Rasa
- Django
- Flask
- Laravel
- Pandas
- Matplotlib
- Seaborn
- Tableau
- Power BI
- RapidMiner
- Datawrapper
- Git & Github
- Bash
- Latex
- Docker
- AWS
Age-related Macular Degeneration (AMD) is a major cause of vision loss. Accurate lesion segmentation aids disease monitoring. This presentation introduces a robust method tailored for retinal images. We use a custom UNet-like architecture with attention and residual blocks. This improves feature capture and segmentation accuracy.
Developed the nation's first Bangla-language contact center AI agent, already deployed for MTB Bank and Land Ministry in Bangladesh, currently undergoing in-house testing at BRAC Bank, Prime Bank, City Bank, LankaBangla and ROBI. The system uses our custom ASR model optimized for Bangla, ensuring high speech recognition accuracy and natural voice interactionsâdelivering a seamless, real-time customer experience, setting a new benchmark for AI-driven service for the Bangladeshi market.
An automated system for Bengali license plate Detection & Recognition. The system uses object detection, OCR, and data archival. It showcases real-time processing and language-specific OCR integration.
The analysis of air quality reveals both improvements and areas of concern. Declines in PM10 and Oâ reflect successful initiatives, but high CO, PM10 and PM2.5 levels pose health risks. Continued efforts are needed to reduce CO sources, control PM2.5 during peak months, and sustain pollution measures.
Developed a Streamlit application that parses PDF resumes into structured JSONâextracting fields such as education, skills, experience, and projectsâand leverages an LLM (Ollama) to predict optimal department matches, streamlining candidate screening.
Developed a CNN-based human face detection model using a diverse dataset of facial images, incorporating various ages, ethnicities, and profiles, with annotated bounding boxes for training and evaluation.
Machine vision was used to classify two raisin types using image features and ML models.
Machine Learning models to predict heart disease based on patient data