Final-Year Software Engineering Student | JavaScript, Python, C++, SQL | Software Development & Problem Solving
I am a final-year Software Engineering student with a strong interest in software development and problem solving. I have built academic and personal projects using Java, Python, C++, SQL, and web technologies. I am looking for an internship or entry-level opportunity where I can apply my technical skills, learn from experienced developers, and contribute to real-world software projects.
FlyRank AI
Aug 2026 — Present· 1 mo
• Build Python data-processing pipelines for preprocessing, feature engineering, model development, and evaluation.
• Analyze real-world search data using machine learning and data science techniques to identify patterns and generate actionable insights.
• Apply engineering methods to AI and search-related problems by building and evaluating data-driven solutions.
CODETREE
May 2025 — Aug 2025· 3 mos
• Diagnosed and resolved hardware, software, and network issues, including LAN configuration and connectivity troubleshooting.
• Supported staff systems and day-to-day IT operations alongside senior engineers.
• Maintained routine system upkeep to ensure smooth IT operations.
2023 – Present
BSc · Computer Science and Engineering · 3.51
2018 – 2020
HSC · Science · 4.50
A full-stack platform for digitizing Dhaka's waste-management workflow from household collection to landfill.
Built a full-stack waste-management platform connecting the workflow from household waste collection to landfill. The system includes role-based dashboards for municipal staff and uses a Next.js frontend, FastAPI backend, and PostgreSQL database. Also implemented a 119-test suite, GitHub Actions CI, and one-command Docker deployment.
A bilingual AI-assisted healthcare platform for preliminary risk assessment, doctor matching, and secure healthcare support.
Built a bilingual Bangla/Banglish/English healthcare platform providing preliminary risk assessment, red-flag detection, BMDC-verified doctor matching, and secure e-prescription support. Developed a 150+ module FastAPI and PostgreSQL backend with 10 deterministic red-flag rules, calibrated ML classification, and FHIR R4 interoperability.
A hybrid AI pipeline for detecting factual hallucinations in Bengali LLM-generated content.
Built a hybrid hallucination-detection pipeline for Bengali LLM outputs. The system combines rule-based checks, Wikipedia retrieval, NLI using mDeBERTa, semantic similarity using LaBSE, and quantized LLM judges including Qwen2.5-7B and Gemma through llama.cpp. The pipeline was optimized using Macro-F1 to detect factually incorrect content in low-resource Bengali outputs.
A real-time AI system for detecting and classifying vehicles in Bangladesh highway surveillance footage.
Built a real-time vehicle detection and classification system for Bangladesh highway surveillance footage. Trained a YOLO11 detector using PyTorch and Ultralytics with a custom augmentation and tuning pipeline, then evaluated the model using mAP@50 and mAP@50–95.
A machine-learning system that predicts customer churn risk and estimates time-to-churn.
Built a survival-analysis system to predict customer churn risk and estimate time-to-churn beyond standard binary classification. Engineered 152 behavioral features from millions of transaction records and developed a calibrated LightGBM, XGBoost, and CatBoost ensemble combined with Lifelines survival models.
2025
Bangladesh Computer Council (ICT Division)
2025
CodeWithHarry
2025
CodeWithHarry