AI Researcher · BSc & MSc in Statistics
· University of Rajshahi · Bangladesh
Statistics taught me to ask the right questions. Machine learning gave me sharper tools. My work connects both: building AI systems efficient enough for edge hardware, rigorous enough for public health decisions, and honest enough about their own failure modes.
Based in Bangladesh, I'm currently working in the Data Science Research Lab of our Statistics Department at the University of Rajshahi, under the supervision of Prof. Dr. M. Rezaul Karim. Now, I'm seeking a PhD position in artificial intelligence, machine learning, or related fields.
, Md. Rezaul Karim
Width should grow 2.8× faster than depth in transformers, validated across 30 architectures up to 7B parameters. Past a critical depth, adding layers actively hurts, even though parameter count increases.
, Md Jahid Hasan Imran , Md. Naim Molla , Luknath Debnath , Tonmoy Shil , Ehsanul Bashar Pranto , Md Mostafizur Rahman Likhon , Md Shafin Sanyan Saad , Md. Rezaul Karim
The revised preprint maps all 17 SDGs across 114 countries and finds 84 supported directed linkages after false-discovery control. No single goal behaves as a stable universal accelerator, so portfolio policy should track time-lagged, evidence-backed linkages instead of rankings alone.
, Soyda Humyra Yesmin , Saiful Islam , Md. Palash Bin Faruque , Md. A. Salam , Md. Mahfuz Uddin , Samiul Islam , Tofayel Ahmed , Md. Binyamin , Md. Rezaul Karim
239× fewer parameters than GraphCast at near-identical accuracy. Principled multi-objective NAS can find truly deployable models, and transfer learning adds ~5% accuracy gains when historical data is scarce.
IEEE 2nd International Conference on Quantum Photonics, Artificial Intelligence & Networking
Apr 2026 - CUET, Chattogram, Bangladesh
“Green-NAS: A Global-Scale Multi-Objective Neural Architecture Search for Robust and Efficient Edge-Native Weather Forecasting”
Regional Conference on Role of Statistics in Strengthening Sustainable Agriculture and Public Health
Apr 2026 - Bangladesh Agricultural University, Mymensingh
“Green-NAS: Multi-Objective Neural Architecture Search for Sustainable and Efficient Weather Forecasting in Data-Scarce Agricultural Regions”
2nd International Conference on Recent Advances in Science and Technology
Nov 2025 - University of Rajshahi, Bangladesh
“Dependency Divide: An Interpretable ML Framework for Profiling Student Digital Satisfaction”
International Conference on Applied Statistics and Data Science
Dec 2025 - University of Dhaka, Bangladesh
“Targeting Renewable Energy Investments to Maximize Emissions Reductions: Precision Climate Finance”
RUEC 1st International Research Conference
Sep 2025 - University of Rajshahi, Bangladesh
“Antibiotic Misuse and Antimicrobial Resistance Knowledge Among University Students in Bangladesh: Current Trends and Future Prospects”
8th International Conference on The Role of Statistics and Data Science in 4IR
Dec 2024 - University of Rajshahi, Bangladesh
Created conference Logo, Banner, Souvenir, and Kit design, alongside technical contributions.
Multi-objective NAS for efficiency under real deployment constraints
A research project on multi-objective neural architecture search (NAS) that explicitly targets efficiency (parameters/energy) while keeping forecasting accuracy competitive.
Evidence that width should grow faster than depth for transformers
A research project studying architecture-conditioned scaling laws and identifying a critical-depth regime where adding layers can hurt performance.
Legacy Fiverr track: 200+ projects, 4.8+ rating, Level Two Seller
WordPress, Shopify, and Wix development for businesses worldwide. This is the older Fiverr lane that established delivery credibility before the AI/data portfolio.
Research Associate
Apr 2025 – PresentData Science Research Lab · University of Rajshahi, Department of Statistics
Working with Prof. M. Rezaul Karim on machine learning applications for public health, neural architecture search, and causal inference for sustainable development goals. Author of 6 arXiv preprints covering ML, transformers, NAS, and statistical modelling.
Level Two Seller, Web Development, Data & Machine Learning
Feb 2021 – Jul 2025200+ successful projects in WordPress, Shopify, Wix, data analysis, and machine learning support. Maintained a 4.8+ rating across 4+ years, spanning custom themes, e-commerce, CMS migration, responsive design, and applied data/ML workflows.
Joint Secretary
Nov 2022 – Dec 2023Nobojagoron Foundation · Rajshahi, Bangladesh
Managed operations for the university's largest volunteer organization (SDG Goal 4). Oversaw volunteer development workshops and skill-building events.
Program Intern
Sep 2022 – Oct 2022Real Star Society · Rajshahi, Bangladesh
Contributed across community development areas: event management, health advocacy campaigns, environmental conservation, and visual content for outreach.
Design Team Lead
Jul 2020 – Oct 2022Nobojagoron Foundation · Rajshahi, Bangladesh
Led all visual communications and campaign design for the foundation's outreach and community programs.
Head of Research & Development
Jan 2020 – Jun 2020Nobojagoron Foundation · Rajshahi, Bangladesh
Volunteer
Jan 2019 – Dec 2019Nobojagoron Foundation · Rajshahi, Bangladesh
Helped underprivileged children with education support, school supplies, and clothing as part of the largest volunteer organization at the University of Rajshahi.
The most expensive equations in AI determine how labs spend billions. Here's what they actually say, and where they're being rewritten. From Kaplan to Chinchilla to inference-time scaling.
Your 7B model has 14 billion numbers. Here's exactly how to shrink them, and what you lose in the process. A practitioner's guide to choosing GGUF, GPTQ, or AWQ.
