Results-driven Artificial Intelligence Engineer and published researcher with a strong background in computer science and applied machine learning. Demonstrated expertise in developing enterprise-grade AI infrastructure, scalable log analytics pipelines, and clinical NLP systems across air-gapped defense environments, fintech, and digital health. Co-author of 15 peer-reviewed publications across IEEE and ACM venues, combining rigorous academic research with production-level model deployment.
Work
Information Technology Directorate, Bangladesh Army
|
Artificial Intelligence Engineer
Dhaka, Dhaka Division, Bangladesh
→
Summary
Architected and delivered secure, air-gapped enterprise AI systems and automated operational workflows across defense infrastructure.
Highlights
Architected and implemented secure, air-gapped, enterprise-grade AI infrastructure and applications to automate high-security organizational workflows.
Designed compartmentalized LLM serving pipelines ensuring 100% data sovereignty and zero external exposure for mission-critical operations.
Collaborated with security and operations divisions to evaluate compliance, system fault tolerance, and low-latency inference requirements.
Periscope Labs Limited
|
Artificial Intelligence Engineer
Dhaka, Dhaka Division, Bangladesh
→
Summary
Engineered AI-driven distributed log analytics pipelines and production ML decision systems for financial and enterprise domains.
Highlights
Designed and implemented AI-driven solutions for automated server log analysis by developing scalable anomaly detection pipelines and LLM workflows for pattern recognition.
Contributed to the development of a machine learning-based Loan Approval System for a leading banking institution, achieving over 80% prediction accuracy via XGBoost and CatBoost.
Engineered robust feature preprocessing pipelines spanning 20,000+ financial records and integrated risk assessment models with FastAPI backend services.
Collaborated with cross-functional engineering teams to integrate, deploy, and evaluate AI models within enterprise systems.
Quantum Robotics and Automation Research Group (QRARG)
|
Junior Researcher
Dhaka, Dhaka Division, Bangladesh
→
Summary
Conducted cross-disciplinary research on intelligent automation, robotics navigation, and natural language understanding models.
Highlights
Collaborated within a multidisciplinary research team to develop novel algorithms for robotics navigation, assistive technologies, and automation systems.
Prepared technical reports, literature reviews, and benchmark evaluations contributing to multiple IEEE conference and journal publications.
Developed prototypes integrating embedded hardware, sensor arrays, and deep learning algorithms for real-time assistive devices.
Military Institute of Science and Technology (MIST)
|
Research Assistant
Dhaka, Dhaka Division, Bangladesh
→
Summary
Conducted applied machine learning and mHealth research on funded institutional healthcare initiatives under academic faculty supervision.
Highlights
Conducted applied research in machine learning and mobile health (mHealth) as part of the PPD Coach project, a MIST-funded initiative supervised by Lt Col Muhammad Nazrul Islam, PhD.
Co-authored multiple peer-reviewed journal and conference papers focusing on diagnostic AI, assistive technologies, and natural language processing.
Supervised and supported the development, testing, and validation of 6+ undergraduate and postgraduate thesis projects focusing on AI, NLP, and Computer Vision.
Robi Axiata Limited
|
Industrial Attachment Intern
Dhaka, Dhaka Division, Bangladesh
→
Summary
Completed comprehensive technical training across telecom core networks, IT infrastructure, and transmission engineering operations.
Highlights
Completed industrial attachment across core telecom domains, including Radio Access Networks (RAN), Core Network, Transport, and IT Infrastructure.
Served as team lead among interns, directing project coordination, sprint reporting, and departmental technical presentations.
Gained hands-on operational exposure to carrier-grade cellular switching, high-availability server workflows, and mobile network operations.
Volunteer
Vice President - Designing & Editing Team
|
MIST Cyber Security Club
→
Directed media branding strategy and assisted in organizing Bangladesh's flagship collegiate cybersecurity conference.
Highlights
Oversaw content creation, brand design strategy, and promotional outreach campaigns across digital club platforms.
Organized and volunteered for LeetCon 2023: HackMeIfYouCan, the 1st International Cyber Security Conference in Bangladesh, facilitating 300+ participants.
Executive Member - Promotion and Design Team
|
MIST Computer Club
→
Produced visual marketing materials and technical project demonstrations for departmental tech exhibits.
Highlights
Utilized digital editing suites to produce club presentations, teasers, and technical project highlights for competitive programming contests.
Team Coordinator - Video Editing Team
|
MIST Innovation Club
→
Led digital media creation and promotional campaigns for institutional innovation initiatives.
Highlights
Led production of promotional, event, and branding videos to support university-wide innovation competitions and showcases.
Designed multimedia assets, introductory animations, and visual content resulting in increased engagement across student platforms.
Education
Military Institute of Science and Technology
Dhaka, Dhaka Division, Bangladesh
→
Bachelor of ScienceComputer Science and Engineering (CSE)Grade: 3.82/4.00
Secondary School Certificate (SSC)ScienceGrade: 5.00/5.00
Awards
1st Runner-Up - BEAR Summit 2025 (AI Category)
World Bank, ICT Division, and BIDA
Recognized among 300+ innovative submissions at Bangladesh's premier deep-tech summit for developing Artificial Kinetic Intelligence (AKI), an AutoML web platform.
Best Paper Award - ICISET 2024
International Conference on Innovations in Science, Engineering and Technology (IEEE)
Honored in the Data Science and Big Data Analytics Track for publication on AI-NLP interactive clinical prescreening systems.
18th Place - ML Olympiad 2024 (Kaggle Challenge)
Kaggle / ML Olympiad Community
Ranked among the top 20 teams nationwide in a machine learning predictive challenge addressing CO2 emissions modeling using real-world environmental datasets.
MIST Dean's List of Honor (2022 & 2023)
Military Institute of Science and Technology
Awarded academic honors and University Merit Scholarships for continuous top-tier academic standing in Computer Science and Engineering.
15th Place - BUET Inter-University Capture the Flag (CTF)
BUET
Secured 15th place in a competitive national collegiate cybersecurity challenge covering web exploitation, forensics, and reverse engineering.
Top 10 Finalist - BUET CSE Fest Hackathon 2023
Bangladesh University of Engineering and Technology (BUET)
Ranked in the top 10 engineering teams in a rapid 24-hour inter-university hackathon designing solutions for real-world constraints.
Microsoft for Startups Founders Hub Acceptance and Azure Grant
Microsoft Corporation
Selected for Microsoft's startup acceleration program, awarded $1,000 in Azure credits and $2,500 in OpenAI credits, followed by a second tier award of $5,000 in Azure credits and specialized Azure architecture guidance.
Education Board Scholarship (SSC & HSC)
Ministry of Education, Bangladesh
Awarded competitive merit scholarship by the Government of Bangladesh for attaining perfect GPA scores in national secondary and higher secondary examinations.
Publications
Sentinel: A multi-pathway architecture for detecting living-off-the-land apt attacks on windows command lines
2026 IEEE Military Communications Conference (MILCOM), IEEE
Investigated advanced persistent threat detection using command-line telemetry and multi-pathway neural network architectures under military communications environments.
A hybrid deep learning model for sentiment analysis of multilingual comments on trending youtube videos
2025 International Conference on Electrical, Computer and Communication Engineering (ECCE), IEEE
Developed a hybrid neural network architecture to process noisy, code-mixed multilingual social media data and infer high-granularity sentiment.
Towards designing a medication assistive intelligent system with automated insulin-dose prediction for elderlies
2025 International Conference on Electrical, Computer and Communication Engineering (ECCE), IEEE
Engineered an assistive predictive healthcare framework to automate personalized insulin dose calculations for geriatric care.
A multi-modal assistive robot navigation system for physically impaired users
2025 International Conference on Quantum Photonics, Artificial Intelligence, and Networking (QPAIN), IEEE
Formulated a multi-modal robotic control and path-planning system leveraging computer vision and sensor fusion for physically disabled users.
Predictive application for pre-screening and analysis of brain tumor using mri
2025 International Conference on Electrical, Computer and Communication Engineering (ECCE), IEEE
Trained convolutional architectures to classify and pre-screen pathological brain anomalies from clinical magnetic resonance imaging scans.
Ai integrated dialogic alphabet zone board for kids in pre-childhood education
2025 International Conference on Electrical, Computer and Communication Engineering (ECCE), IEEE
Designed an interactive hardware-software educational interface incorporating speech processing to assist early childhood linguistic acquisition.
Exploring the usability of vending machines through multi-method evaluation techniques
2025 International Conference on Electrical, Computer and Communication Engineering (ECCE), IEEE
Conducted a structured human-computer interaction study analyzing user experience, physical interface bottlenecks, and transactional efficiency.
Azure health bot based support system for personal healthcare and monitoring
Proceedings of the 3rd International Conference on Computing Advancements (ICCA '24), ACM
Constructed an automated conversational health monitoring application leveraging cloud microservices and medical triage conversational models.
Bilingual sms spam detection using deep ensemble learning
2024 IEEE International Women in Engineering Conference on Electrical and Computer Engineering (WIECON-ECE), IEEE
Proposed a multi-model deep ensemble architecture to classify bilingual SMS messages with robust accuracy against adversarial tokens.
Proposing and evaluating a modified keyboard layout for one-handed users
2024 27th International Conference on Computer and Information Technology (ICCIT), IEEE
Designed and evaluated an ergonomic one-handed input layout optimizing keystroke distribution and typing velocity for motor-impaired individuals.
Modeling a fuzzy expert system for predicting customers' behavioral intention for restaurant business through parallel prediction of perceived values and satisfactions
International Journal of Information Technology, Springer
Developed a fuzzy expert system utilizing parallel value-satisfaction modeling to accurately forecast consumer behavioral patterns. Quartile: Q2, IF: 3.07.
Advancing bilingual sms spam detection: A comparative exploration of machine learning and deep ensemble techniques
2024 IEEE 3rd International Conference on Robotics, Automation, Artificial-Intelligence and Internet-of-Things (RAAICON), IEEE
Evaluated classical classification models against deep neural ensembles for cross-lingual English and Bengali spam text filtering.
An ai-nlp based interactive chatbot model for patient prescreening at doctor's consultancy
2024 International Conference on Innovations in Science, Engineering and Technology (ICISET), IEEE
Constructed an interactive natural language processing dialogue agent to triage patient symptoms prior to clinical consultations (Awarded Best Paper).
Application of phi (phi), the golden ratio, in computing: A systematic review
IEEE Access
Comprehensive systematic literature review examining mathematical convergence and golden ratio optimization across computational algorithms. Quartile: Q1, IF: 3.4.
Rouge score analysis and performance evaluation between google t5 and spacy for youtube news video summarization
2023 26th International Conference on Computer and Information Technology (ICCIT), IEEE
Conducted quantitative comparative evaluations between abstractive transformer (T5) and extractive statistical (SpaCy) summarization pipelines.