🎓 Bachelor of Computer Science & Engineering (BCSE) student at the International University of Business Agriculture and Technology (IUBAT).

🤖 I am passionate about Artificial Intelligence, Machine Learning, Data Science, Computer Vision, Natural Language Processing (NLP), and Large Language Models (LLMs).

💡 I enjoy building AI-powered applications, participating in hackathons, and solving real-world problems through machine learning and deep learning.

💡 I enjoy building AI-powered applications, participating in hackathons, solving real-world problems through machine learning, and continuously exploring emerging AI technologies.

🎯 Career Goal: To build impactful AI systems and contribute to research and industry by developing intelligent, scalable, and data-driven solutions.

• Artificial Intelligence (AI)
• Machine Learning (ML)
• Deep Learning
• Large Language Models (LLMs)
• Natural Language Processing (NLP)
• Computer Vision
• Time Series Forecasting
• Financial Risk Analytics
• Retrieval-Augmented Generation (RAG)

Libraries & Frameworks

• Scikit-learn
• LightGBM
• XGBoost
• YOLO
• LangChain
• llama.cpp
• Gemma

• Jupyter Notebook
• Google Colab
• Kaggle

Hybrid Bengali hallucination detection using Rule-based Validation, Retrieval techniques, and Large Language Models (LLMs).

🔗 https://github.com/meherabmehu/Bengali-Hallucination-Detection

Developed a survival analysis pipeline using 152 engineered behavioral features extracted from millions of financial transactions.

🔗 https://github.com/meherabmehu/Customer-Churn-Prediction-using-Survival-Analysis-

YOLO-based traffic detection and vehicle classification system developed for Bangladesh highway surveillance imagery.

🔗 https://github.com/meherabmehu/RoadVision-AI-Powered-Traffic-Detection-Vehicle-Classification-

Deep learning-based forecasting model using LSTM, GRU, and ARIMA for 24-hour ahead PM2.5 prediction.

🔗 https://github.com/meherabmehu/PM2.5-Air-Quality-Forecasting-for-Dhaka-

Machine learning solution for detecting risky financial transactions using anonymized behavioral features.

🔗 https://github.com/meherabmehu/finnova-risk-transaction-detection

AI-powered binary classification solution developed for the NDSS 2026 Data Hackathon.

🔗 https://github.com/meherabmehu/Starship-Safety-NDSS-2026

• 🥈 1st Runner-Up — 5th Data Science Summit 2026 (Data Hackathon), Daffodil International University (DIU)
• 🥈 First Runner-Up — bKash × NSUCEC Cybernauts 2026 Datathon Grand Finale
• 🥈 First Runner-Up — DUET CSE Carnival 2026 AI Hackathon

To build impactful AI systems and contribute to research and industry by developing intelligent, scalable, and data-driven solutions.

📧 Email: jamimmhj@gmail.com

💼 LinkedIn:
https://www.linkedin.com/in/meherab-talukder-5046a141b/

📊 Kaggle:
https://www.kaggle.com/mdmeherabhossain

🐙 GitHub:
https://github.com/meherabmehu

⭐ Thank you for visiting my GitHub profile! Feel free to explore my repositories and connect with me.