AI Researcher · NLP · Computer Vision · Graph Learning · Trustworthy Multimodal AI

Building reliable, explainable, and human-centered AI systems for low-resource languages, document understanding, and real-world intelligent applications.

I am an AI researcher from Dhaka, Bangladesh, working across natural language processing, computer vision, graph learning, and trustworthy multimodal systems. My research interests include long-context reasoning, diagram and document understanding, Bangla and low-resource NLP, graph-based learning, and safety-aware AI evaluation.

Alongside research, I work on practical AI systems that connect academic ideas with deployable products. I am also committed to inclusive technology education in Bangladesh and have helped more than 2,000 learners explore computing, AI, and digital literacy.

• Vision-language model reliability: grounding multimodal reasoning to reduce structural and topology hallucinations.
• Bangla and low-resource NLP: building datasets, benchmarks, and models for summarization, translation, and cultural knowledge evaluation.
• Graph-based AI systems: using graph structure, reasoning pipelines, and explainability for robust decision-making.
• Applied AI deployment: turning research-backed AI into scalable tools for education, business, and public-interest use cases.

B.Sc. in Computer Science and Engineering
January 2020 — June 2025 · Dhaka, Bangladesh
CGPA: 3.90/4.00 · Top of Class

Higher Secondary School Certificate, Science
June 2017 — July 2019 · Dhaka, Bangladesh
GPA: 5.00/5.00

Secondary School Certificate, Science
January 2015 — May 2017 · Noakhali, Bangladesh
GPA: 5.00/5.00

Dean's Award for Academic Excellence Institution: Faculty of Engineering & Technology, University of Dhaka Recognition: Top academic performance in Computer Science & Engineering Context: CGPA 3.90/4.00 · Top of Class

Inter-University NLP Idea Competition Linguistics Department, University of Dhaka

Bhashamul: Inter-University NLP Competition United International University

DL Enigma 1.0: Deep Learning Competition Shahjalal University of Science & Technology

Code Samurai 2024 · Final Round University of Dhaka

Code Samurai 2022 · Final Round University of Dhaka

Best Volunteer Award Dhaka University IT Society Annual Function

Bangladesh Physics Olympiad · Dhaka North Regional Round Higher Secondary View certificate

Bangladesh Physics Olympiad · National Round Junior View certificate

Bangladesh Mathematical Olympiad · Noakhali Region Primary View certificate

ICDAR 2026 · 20th International Conference on Document Analysis and Recognition
Authors: Abdullah Ibne Hanif Arean, Niamul Hassan Samin, Md Arifur Rahman, Renu Akter Suity, Juena Ahmed Noshin, Md Ashikur Rahman

• Introduces Stroke-Level Connectivity Verification (SLCV) to ground diagram reasoning in continuous pixel-level stroke paths.
• Proposes PaRCO, a modular architecture that separates structural reconstruction, semantic reasoning, and constraint validation.
• Frames topology-aware verification as a reusable primitive for reliable multimodal document understanding.

AACL-IJCNLP 2025 Findings · OpenReview
Authors: Fariha Anjum Shifa, Muhtasim Ibteda Shochcho, Abdullah Ibne Hanif Arean, Mohammad Ashfaq Ur Rahman, Akm Moshiur Rahman Mazumder, Ahaj Faiak, Md Fahim, M. Ashraful Amin, Amin Ahsan Ali, AKM Mahbubur Rahman

• Introduces a 4,234-question Bangla benchmark for evaluating cultural and social knowledge in LLMs.
• Benchmarks open-source and closed-source models and analyzes remaining gaps after LoRA fine-tuning.
• Expands Bangla LLM evaluation beyond STEM-heavy datasets toward local culture, society, and commonsense knowledge.

arXiv Preprint · February 2025
Authors: Istiaq Ahmed Fahad, Abdullah Ibne Hanif Arean, Nazmus Sakib Ahmed, Mahmudul Hasan

• Develops a DETR-based framework for automatic vehicle detection in complex road, lighting, and weather conditions.
• Uses Co-DETR with collaborative hybrid assignment training to improve feature learning and training supervision.
• Compares DETR variants and YOLO baselines on BadODD to study robustness for real-world autonomous navigation.

June 2025 — Present · Dhaka, Bangladesh · Full-time, On-site

• Developing graph-based reasoning pipelines to improve multimodal interpretation in vision-language models.
• Leading research on specialized VLMs for diagram understanding, with focus on structural grounding and hallucination reduction.
• Applying explainable AI principles to support deployment in safety-critical and high-reliability domains.

July 2025 — Present · United States · Remote, Part-time

• Co-leading the development of enterprise-focused AI systems with emphasis on measurable business impact.
• Overseeing production-ready AI deployment pipelines for scalable integration into real-world workflows.
• Building research-backed AI, NLP, and automation systems for business decision-making and operational efficiency.

December 2025 — Present · Dhaka, Bangladesh · Part-time

• Advising digital transformation initiatives for EIIN-registered schools and colleges through Amar IMS.
• Supporting SaaS-based solutions for academic administration, data collection, payment management, and operational efficiency.
• Providing technical and strategic guidance for scalable, secure, and cost-effective IT deployment.

January 2024 — May 2025 · Dhaka, Bangladesh · Full-time

• Collected and processed Bangla-language datasets containing more than 50 million sentences.
• Built LLM-based systems for Bangla summarization, translation, and related NLP tasks.
• Contributed to Bangla handwritten image recognition research and dataset preparation.

April 2025 — May 2025 · United States · Remote, Part-time

• Developed AI-driven tools using large language models and modern automation techniques.
• Collaborated with engineers on validation, verification, and workflow automation.

June 2024 — March 2025 · Web Application · Presentation · Thesis

• Built a Bangla news aggregation platform that ingests, normalizes, and summarizes articles from multiple publishers.
• Designed LongBanSum, a T5-based long-sequence summarization model using local-global attention.
• Curated MASBA, a 669,699-article Bangla corpus with multi-level human-written summaries.

January 2023 — June 2023

• Developed a centralized portal for on-demand IT support and common technical services.
• Led service-flow design, technical documentation, and release coordination.
• Implemented backend services using Spring Boot, Spring MVC, Java, authentication, and RESTful APIs.

July 2022 — November 2022 · Presentation

• Developed an Android application in Java for disaster relief distribution and resource tracking.
• Designed a secure backend API and SQLite schema for reliable updates, data integrity, and offline resilience.

May 2025 — April 2026 · Dhaka, Bangladesh

• Leading student initiatives that promote digital literacy, practical computing, and technology awareness.
• Organized and conducted sessions on PC building, academic search skills, and cybersecurity awareness for women.
• Strengthening collaboration among students, alumni, educators, and technology professionals.

August 2024 — March 2025 · Dhaka, Bangladesh

• Collected and analyzed casualty and martyrdom-related data from the July Revolution in collaboration with MIS.
• Served as student representative of Shaheed Suhrawardy Medical College and helped lead the July Shaheed Smrity Foundation Help Center.

January 2024 — February 2025 · Dhaka, Bangladesh

• Led the IEEE Computer Society Student Branch Executive Committee at the University of Dhaka.
• Coordinated technical events, workshops, seminars, and student engagement initiatives.
• Promoted IEEE Computer Society activities and collaboration within the academic community.

I am open to research and engineering collaborations in:

• Bangla and low-resource NLP
• Multilingual LLM safety and evaluation
• Vision-language model grounding and hallucination reduction
• Diagram, document, and structured visual reasoning
• Graph learning and explainable AI
• AI for education, public-interest technology, and inclusive digital access