CSE graduate from Dhaka who builds machine learning models and the apps around them — outbreak risk, grid faults, campaign KPIs — and ships them as live, interactive tools.
I'm a Computer Science and Engineering graduate from North South University working across the full stack — C#, JavaScript, and Python machine learning pipelines.
I care about the whole model lifecycle: feature engineering, training with TensorFlow, PyTorch and XGBoost, honest evaluation against baselines, and deployment as a Streamlit app someone can actually use.
B.Sc. Computer Science and Engineering
Dhaka, BangladeshHigher Secondary Certificate, Science
2018 · Cumilla
Multi-output XGBoost regression forecasting 6 campaign KPIs — conversion rate, acquisition cost, ROI, clicks, impressions, engagement — from 8 campaign attributes, compared side by side with historical actuals.
Fault diagnosis on the IEEE 14-bus system: classical ML benchmarked against a deep neural network for early-stage grid fault detection.
University research project: 1,341 records of admissions, deaths and weather used to classify outbreak risk, with 5 models compared and Random Forest the strongest. Deployed as a live risk tool for Bangladeshi cities.
Recommendation system for fast-moving consumer goods that fuses text descriptions, user metadata and product imagery into personalized suggestions.
