CV
Curriculum vitae of Shadman Sakib — software engineer and aspiring researcher.
Contact Information
| Name | Shadman Sakib |
| Professional Title | Software Engineer |
| shadmansakib20@iut-dhaka.edu | |
| Phone | +880 1608-637421 |
| Location | Dhaka, Bangladesh |
| Website | https://sakibshadman.com |
Professional Summary
Software Engineer specializing in backend development (Java, Spring Boot) for enterprise-scale government systems, with research focused on automated software verification and model conformance, large language models for requirements engineering, and trustworthy AI.
Experience
-
Oct 2025 - Present Dhaka, Bangladesh
Software Engineer I
Synesis IT PLC
- Backend development in Java and Spring Boot for enterprise-scale government information systems, working directly with real-world software complexity, legacy codebase maintenance, and long-term system evolution.
- Refactored legacy modules to reduce technical debt and improve structural consistency across large, distributed codebases.
- Designed JWT-based authentication and authorisation across distributed microservices; developed RESTful APIs serving high-volume production traffic.
-
Jun 2024 - Oct 2024 Dhaka, Bangladesh
Backend Developer Intern
Synesis IT PLC
- Contributed to the national E-TIN tax platform backend, integrating authentication workflows and improving service modularity.
- Participated in a full Agile/Scrum development lifecycle across a multi-team engineering organisation.
Education
-
2021 - 2025 Dhaka, Bangladesh
B.Sc. in Software Engineering
Islamic University of Technology (IUT)
Software Engineering
- CGPA: 3.26 / 4.00
- Thesis: LLM-Based Requirements Generation from User Reviews — designed an agentic, prompt-driven pipeline to extract structured functional requirements (user stories) from unstructured app-store reviews.
- Evaluated machine-generated requirements against human-written baselines using semantic similarity and structural validity metrics, identifying systematic conformance gaps and failure modes.
Academic Interests
Skills
Projects
-
Automated Requirements Conformance Evaluation
B.Sc. thesis extension on structural conformance of LLM-generated requirements.
- Systematically classified structural conformance failures in LLM-generated user stories, building a taxonomy of deviation types from standard requirement specifications.
- Developed Python evaluation scripts to automate structural and semantic comparison between generated and reference artefacts.
-
Multimodal Emotion Recognition System
Computer vision + NLP pipeline for real-time emotion prediction.
- Built a multimodal pipeline fusing CNN-based facial feature extraction with NLP text-sentiment embeddings for real-time emotion prediction.
- Applied preprocessing optimisations and hyperparameter tuning to improve cross-modal fusion performance.
Publications
-
2026 From Reviews to Requirements: Can LLMs Generate Human-Like User Stories?
NLP4RE'26 @ REFSQ (Accepted)
Proposed an automated pipeline generating structured software requirements from app-store user feedback, evaluating prompting strategies and categorising failure modes in LLM-generated requirements.
-
2025 From Chat to Checkup: Can Large Language Models Assist in Diabetes Prediction?
IEEE Xplore (Published)
Benchmarked LLM-based inference against traditional ML classifiers on the Pima Indians Diabetes Dataset, analysing reliability, interpretability, and failure conditions of generative models in a safety-critical context.
Academic Activities
- IUT Mars Rover Anirban — Documentation Team (2022–2023): Produced technical documentation and system reports for an international university robotics competition.
- Collaborated within a cross-disciplinary team spanning software, electronics, and mechanical sub-teams.